MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications

Abstract

We present a class of efficient models called MobileNets for mobile and embedded vision applications. MobileNets are based on a streamlined architecture that uses depth-wise separable convolutions to build light weight deep neural networks. We introduce two simple global hyper-parameters that efficiently trade off between latency and accuracy. These hyper-parameters allow the model builder to choose the right sized model for their application based on the constraints of the problem. We present extensive experiments on resource and accuracy tradeoffs and show strong performance compared to other popular models on ImageNet classification. We then demonstrate the effectiveness of MobileNets across a wide range of applications and use cases including object detection, finegrain classification, face attributes and large scale geo-localization.

References in corpus (8)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Distilling the Knowledge in a Neural Network
• Caffe: Convolutional Architecture for Fast Feature Embedding
• Quantized Neural Networks: Training Neural Networks with Low Precision Weights and Activations
• Going Deeper with Convolutions
• Compressing Neural Networks with the Hashing Trick
• Speeding up Convolutional Neural Networks with Low Rank Expansions
• Speeding-up Convolutional Neural Networks Using Fine-tuned CP-Decomposition

Cited by in corpus (2114)

• YOLOv4: Optimal Speed and Accuracy of Object Detection
• EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
• Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation
• Knowledge Distillation: A Survey
• Covid-19: Automatic detection from X-Ray images utilizing Transfer Learning with Convolutional Neural Networks
• PVT v2: Improved Baselines with Pyramid Vision Transformer
• MobileNetV2: Inverted Residuals and Linear Bottlenecks
• Conv-TasNet: Surpassing Ideal Time-Frequency Magnitude Masking for Speech Separation
• NTU RGB+D 120: A Large-Scale Benchmark for 3D Human Activity Understanding
• Convergence of Edge Computing and Deep Learning: A Comprehensive Survey
• Deep Joint Source-Channel Coding for Wireless Image Transmission
• MLP-Mixer: An all-MLP Architecture for Vision
• Deep Multi-modal Object Detection and Semantic Segmentation for Autonomous Driving: Datasets, Methods, and Challenges
• A Survey of Deep Learning-based Object Detection
• The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks
• AMC: AutoML for Model Compression and Acceleration on Mobile Devices
• Activation Functions: Comparison of trends in Practice and Research for Deep Learning
• Deep Face Recognition: A Survey
• ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
• Road Damage Detection Using Deep Neural Networks with Images Captured Through a Smartphone
• A Survey of Model Compression and Acceleration for Deep Neural Networks
• Benchmark Analysis of Representative Deep Neural Network Architectures
• Medical Image Segmentation Using Deep Learning: A Survey
• Deep neural network models for computational histopathology: A survey
• Searching for Activation Functions
• PlantDoc: A Dataset for Visual Plant Disease Detection
• Regularized Deep Networks in Intelligent Transportation Systems: A Taxonomy and a Case Study
• A survey on modern trainable activation functions
• Object Detection in 20 Years: A Survey
• Binary Neural Networks: A Survey
• Gather-Excite: Exploiting Feature Context in Convolutional Neural Networks
• A Review on Deep Learning in UAV Remote Sensing
• Monocular Human Pose Estimation: A Survey of Deep Learning-based Methods
• Asynchronous Federated Optimization
• RetinaFace: Single-stage Dense Face Localisation in the Wild
• PC-DARTS: Partial Channel Connections for Memory-Efficient Architecture Search
• Avoiding Overfitting: A Survey on Regularization Methods for Convolutional Neural Networks
• Do ImageNet Classifiers Generalize to ImageNet?
• Advancements in Image Classification using Convolutional Neural Network
• Visual Transformers: Token-based Image Representation and Processing for Computer Vision
• Deep Learning for UAV-based Object Detection and Tracking: A Survey
• A Survey: Deep Learning for Hyperspectral Image Classification with Few Labeled Samples
• Fast-SCNN: Fast Semantic Segmentation Network
• CSPNet: A New Backbone that can Enhance Learning Capability of CNN
• Extracting possibly representative COVID-19 Biomarkers from X-Ray images with Deep Learning approach and image data related to Pulmonary Diseases
• Hello Edge: Keyword Spotting on Microcontrollers
• Bayesian Compression for Deep Learning
• Point-Voxel CNN for Efficient 3D Deep Learning
• Coordinate Attention for Efficient Mobile Network Design
• Training and Inference with Integers in Deep Neural Networks
• Wide Activation for Efficient and Accurate Image Super-Resolution
• MixConv: Mixed Depthwise Convolutional Kernels
• Rice Diseases Detection and Classification Using Attention Based Neural Network and Bayesian Optimization
• Split Computing and Early Exiting for Deep Learning Applications: Survey and Research Challenges
• LocalViT: Analyzing Locality in Vision Transformers
• ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware
• Deep learning-based survival prediction for multiple cancer types using histopathology images
• SNAS: Stochastic Neural Architecture Search
• MCUNet: Tiny Deep Learning on IoT Devices
• NAS-FAS: Static-Dynamic Central Difference Network Search for Face Anti-Spoofing
• Pruning by Explaining: A Novel Criterion for Deep Neural Network Pruning
• A Survey of Convolutional Neural Networks: Analysis, Applications, and Prospects
• DARTS+: Improved Differentiable Architecture Search with Early Stopping
• Face Mask Detection using Transfer Learning of InceptionV3
• Deep Dual-resolution Networks for Real-time and Accurate Semantic Segmentation of Road Scenes
• BlazeFace: Sub-millisecond Neural Face Detection on Mobile GPUs
• Depthwise Separable Convolutions for Neural Machine Translation
• Modality specific U-Net variants for biomedical image segmentation: A survey
• The impact of patient clinical information on automated skin cancer detection
• Slalom: Fast, Verifiable and Private Execution of Neural Networks in Trusted Hardware
• ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design
• Single Path One-Shot Neural Architecture Search with Uniform Sampling
• Slimmable Neural Networks
• Deep Learning for Generic Object Detection: A Survey
• Interpretable Survival Prediction for Colorectal Cancer using Deep Learning
• SimpleShot: Revisiting Nearest-Neighbor Classification for Few-Shot Learning
• Benchmarking TPU, GPU, and CPU Platforms for Deep Learning
• Hyperspectral Classification Based on Lightweight 3-D-CNN With Transfer Learning
• Federated Learning in the Sky: Aerial-Ground Air Quality Sensing Framework with UAV Swarms
• Cloudburst: Stateful Functions-as-a-Service
• FastFCN: Rethinking Dilated Convolution in the Backbone for Semantic Segmentation
• EmergencyNet: Efficient Aerial Image Classification for Drone-Based Emergency Monitoring Using Atrous Convolutional Feature Fusion
• HAT: Hardware-Aware Transformers for Efficient Natural Language Processing
• Interstellar: Using Halide's Scheduling Language to Analyze DNN Accelerators
• Stand-Alone Self-Attention in Vision Models
• Integer Quantization for Deep Learning Inference: Principles and Empirical Evaluation
• Hardware and Software Optimizations for Accelerating Deep Neural Networks: Survey of Current Trends, Challenges, and the Road Ahead
• GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyond
• Revisiting ResNets: Improved Training and Scaling Strategies
• Demystifying Parallel and Distributed Deep Learning: An In-Depth Concurrency Analysis
• A Survey on Neural Architecture Search
• Skin Lesion Analyser: An Efficient Seven-Way Multi-Class Skin Cancer Classification Using MobileNet
• GluonCV and GluonNLP: Deep Learning in Computer Vision and Natural Language Processing
• Quantization and Deployment of Deep Neural Networks on Microcontrollers
• A Comprehensive Survey of Neural Architecture Search: Challenges and Solutions
• ContextNet: Exploring Context and Detail for Semantic Segmentation in Real-time
• Benchmarking TinyML Systems: Challenges and Direction
• Automatic heterogeneous quantization of deep neural networks for low-latency inference on the edge for particle detectors
• GhostNets on Heterogeneous Devices via Cheap Operations
• You Only Look at One Sequence: Rethinking Transformer in Vision through Object Detection
• Group Knowledge Transfer: Federated Learning of Large CNNs at the Edge
• DarkneTZ: Towards Model Privacy at the Edge using Trusted Execution Environments
• DABNet: Depth-wise Asymmetric Bottleneck for Real-time Semantic Segmentation
• Learning to Optimize Tensor Programs
• Empowering Edge Intelligence: A Comprehensive Survey on On-Device AI Models
• Music Source Separation in the Waveform Domain
• Tensor Methods in Computer Vision and Deep Learning
• Adversarial Example Detection for DNN Models: A Review and Experimental Comparison
• Neural Architecture Transfer
• Spatial Group-wise Enhance: Improving Semantic Feature Learning in Convolutional Networks
• EEG-based Cross-Subject Driver Drowsiness Recognition with an Interpretable Convolutional Neural Network
• MobileSal: Extremely Efficient RGB-D Salient Object Detection
• Real-time object detection method based on improved YOLOv4-tiny
• L1-Norm Batch Normalization for Efficient Training of Deep Neural Networks
• Towards the Systematic Reporting of the Energy and Carbon Footprints of Machine Learning
• Deep-learning-based surrogate flow modeling and geological parameterization for data assimilation in 3D subsurface flow
• Real-time Pose and Shape Reconstruction of Two Interacting Hands With a Single Depth Camera
• Bringing AI To Edge: From Deep Learning's Perspective
• CvT: Introducing Convolutions to Vision Transformers
• TSM: Temporal Shift Module for Efficient Video Understanding
• Accelerating COVID-19 Differential Diagnosis with Explainable Ultrasound Image Analysis
• Progressive Neural Architecture Search
• DeeperLab: Single-Shot Image Parser
• CornerNet-Lite: Efficient Keypoint Based Object Detection
• Discrimination-aware Channel Pruning for Deep Neural Networks
• Evaluating the Robustness of Neural Networks: An Extreme Value Theory Approach
• ViTAE: Vision Transformer Advanced by Exploring Intrinsic Inductive Bias
• Discrimination-aware Network Pruning for Deep Model Compression
• MobileViT: Light-weight, General-purpose, and Mobile-friendly Vision Transformer
• Label Refinery: Improving ImageNet Classification through Label Progression
• Path-Level Network Transformation for Efficient Architecture Search
• A Comprehensive Overview and Comparative Analysis on Deep Learning Models: CNN, RNN, LSTM, GRU
• Auto-DeepLab: Hierarchical Neural Architecture Search for Semantic Image Segmentation
• Drop an Octave: Reducing Spatial Redundancy in Convolutional Neural Networks with Octave Convolution
• Bag of Freebies for Training Object Detection Neural Networks
• AutoSlim: Towards One-Shot Architecture Search for Channel Numbers
• FETCH: A deep-learning based classifier for fast transient classification
• Bag of Tricks for Image Classification with Convolutional Neural Networks
• How to Prove Your Model Belongs to You: A Blind-Watermark based Framework to Protect Intellectual Property of DNN
• Layer-wise training convolutional neural networks with smaller filters for human activity recognition using wearable sensors
• A Survey of FPGA-Based Neural Network Accelerator
• SiamRPN++: Evolution of Siamese Visual Tracking with Very Deep Networks
• Deep Learning for Automatic Pneumonia Detection
• Enable Deep Learning on Mobile Devices: Methods, Systems, and Applications
• High-Throughput CNN Inference on Embedded ARM big.LITTLE Multi-Core Processors
• TVM: An Automated End-to-End Optimizing Compiler for Deep Learning
• Automatic classification between COVID-19 pneumonia, non-COVID-19 pneumonia, and the healthy on chest X-ray image: combination of data augmentation methods
• AI Challenger : A Large-scale Dataset for Going Deeper in Image Understanding
• Backdoor Attacks and Countermeasures on Deep Learning: A Comprehensive Review
• Progressive Differentiable Architecture Search: Bridging the Depth Gap between Search and Evaluation
• TensorFlow.js: Machine Learning for the Web and Beyond
• Exploring Spatial Significance via Hybrid Pyramidal Graph Network for Vehicle Re-identification
• Compact Generalized Non-local Network
• FastDeepIoT: Towards Understanding and Optimizing Neural Network Execution Time on Mobile and Embedded Devices
• Incremental Learning in Deep Convolutional Neural Networks Using Partial Network Sharing
• Deep Learning in Mobile and Wireless Networking: A Survey
• Tip-Adapter: Training-free CLIP-Adapter for Better Vision-Language Modeling
• Towards Learning a Universal Non-Semantic Representation of Speech
• Edge Intelligence: Architectures, Challenges, and Applications
• Interventional Few-Shot Learning
• TPH-YOLOv5: Improved YOLOv5 Based on Transformer Prediction Head for Object Detection on Drone-captured Scenarios
• Receptive Field Block Net for Accurate and Fast Object Detection
• Compressing Neural Networks using the Variational Information Bottleneck
• RetinaMask: Learning to predict masks improves state-of-the-art single-shot detection for free
• Rigging the Lottery: Making All Tickets Winners
• XNOR-Net++: Improved Binary Neural Networks
• Omni-Scale Feature Learning for Person Re-Identification
• Streaming keyword spotting on mobile devices
• CorrDetector: A Framework for Structural Corrosion Detection from Drone Images using Ensemble Deep Learning
• GhostNet: More Features from Cheap Operations
• BiSeNet V2: Bilateral Network with Guided Aggregation for Real-time Semantic Segmentation
• Combining a Convolutional Neural Network with Autoencoders to Predict the Survival Chance of COVID-19 Patients
• Recent Advances in Deep Learning: An Overview
• DeepCABAC: A Universal Compression Algorithm for Deep Neural Networks
• CLIP-Adapter: Better Vision-Language Models with Feature Adapters
• Machine learning and AI-based approaches for bioactive ligand discovery and GPCR-ligand recognition
• Recent Advances in Object Detection in the Age of Deep Convolutional Neural Networks
• Light-Weight RefineNet for Real-Time Semantic Segmentation
• Rethinking on Multi-Stage Networks for Human Pose Estimation
• Hardware Acceleration of Sparse and Irregular Tensor Computations of ML Models: A Survey and Insights
• Dynamic Fusion Module Evolves Drivable Area and Road Anomaly Detection: A Benchmark and Algorithms
• Review: Deep Learning in Electron Microscopy
• Deep Learning-based Spacecraft Relative Navigation Methods: A Survey
• Adaptive Inference through Early-Exit Networks: Design, Challenges and Directions
• RF-Based Human Activity Recognition Using Signal Adapted Convolutional Neural Network
• Siamese Attentional Keypoint Network for High Performance Visual Tracking
• PointSeg: Real-Time Semantic Segmentation Based on 3D LiDAR Point Cloud
• An Empirical Study of Spatial Attention Mechanisms in Deep Networks
• Face Recognition: From Traditional to Deep Learning Methods
• Data-Free Adversarial Distillation
• CSWin Transformer: A General Vision Transformer Backbone with Cross-Shaped Windows
• Pruning Deep Convolutional Neural Networks Architectures with Evolution Strategy
• Balanced Semi-Supervised Generative Adversarial Network for Damage Assessment from Low-Data Imbalanced-Class Regime
• Joint Device-Edge Inference over Wireless Links with Pruning
• Tiny-DSOD: Lightweight Object Detection for Resource-Restricted Usages
• Pelee: A Real-Time Object Detection System on Mobile Devices
• Soft Threshold Weight Reparameterization for Learnable Sparsity
• Edge Intelligence: Paving the Last Mile of Artificial Intelligence with Edge Computing
• Once-for-All: Train One Network and Specialize it for Efficient Deployment
• CycleMLP: A MLP-like Architecture for Dense Prediction
• ATRW: A Benchmark for Amur Tiger Re-identification in the Wild
• Shallow-Deep Networks: Understanding and Mitigating Network Overthinking
• Efficient Two-Stream Network for Violence Detection Using Separable Convolutional LSTM
• MetaPruning: Meta Learning for Automatic Neural Network Channel Pruning
• PP-LCNet: A Lightweight CPU Convolutional Neural Network
• CondenseNet: An Efficient DenseNet using Learned Group Convolutions
• Score-CAM: Score-Weighted Visual Explanations for Convolutional Neural Networks
• Selective Kernel Networks
• MobileBERT: a Compact Task-Agnostic BERT for Resource-Limited Devices
• Accelerator-Aware Pruning for Convolutional Neural Networks
• The Power of Sparsity in Convolutional Neural Networks
• Deep $k$-Means: Re-Training and Parameter Sharing with Harder Cluster Assignments for Compressing Deep Convolutions
• Revisiting Shadow Detection: A New Benchmark Dataset for Complex World
• Loss-aware Weight Quantization of Deep Networks
• Real-time monitoring of driver drowsiness on mobile platforms using 3D neural networks
• FBNet: Hardware-Aware Efficient ConvNet Design via Differentiable Neural Architecture Search
• What Do Compressed Deep Neural Networks Forget?
• Non-Vacuous Generalization Bounds at the ImageNet Scale: A PAC-Bayesian Compression Approach
• Double Similarity Distillation for Semantic Image Segmentation
• Efficient Facial Representations for Age, Gender and Identity Recognition in Organizing Photo Albums using Multi-output CNN
• Prediction of lung and colon cancer through analysis of histopathological images by utilizing Pre-trained CNN models with visualization of class activation and saliency maps
• Automated interpretation of congenital heart disease from multi-view echocardiograms
• Dynamic Sampling Networks for Efficient Action Recognition in Videos
• Closing the Generalization Gap of Adaptive Gradient Methods in Training Deep Neural Networks
• Measuring the Algorithmic Efficiency of Neural Networks
• MEDIC: A Multi-Task Learning Dataset for Disaster Image Classification
• Paraphrasing Complex Network: Network Compression via Factor Transfer
• Video Classification with Channel-Separated Convolutional Networks
• Visual Wake Words Dataset
• SpArSe: Sparse Architecture Search for CNNs on Resource-Constrained Microcontrollers
• CGNet: A Light-weight Context Guided Network for Semantic Segmentation
• Learning to Decode the Surface Code with a Recurrent, Transformer-Based Neural Network
• SkyNet: a Hardware-Efficient Method for Object Detection and Tracking on Embedded Systems
• Be Your Own Teacher: Improve the Performance of Convolutional Neural Networks via Self Distillation
• Spatiotemporal Contrastive Video Representation Learning
• Stabilizing Differentiable Architecture Search via Perturbation-based Regularization
• Deep Learning Methods for Solving Linear Inverse Problems: Research Directions and Paradigms
• Deep Learning of Quasar Spectra to Discover and Characterize Damped Lya Systems
• Exploring Randomly Wired Neural Networks for Image Recognition
• Ansor: Generating High-Performance Tensor Programs for Deep Learning
• MobiSR: Efficient On-Device Super-Resolution through Heterogeneous Mobile Processors
• Heterogeneous Multilayer Generalized Operational Perceptron
• Deep Learning Inference in Facebook Data Centers: Characterization, Performance Optimizations and Hardware Implications
• Local Motion Planner for Autonomous Navigation in Vineyards with a RGB-D Camera-Based Algorithm and Deep Learning Synergy
• Deep Neural Network Approximation for Custom Hardware: Where We've Been, Where We're Going
• Net2Vis -- A Visual Grammar for Automatically Generating Publication-Tailored CNN Architecture Visualizations
• Network Decoupling: From Regular to Depthwise Separable Convolutions
• Applications and Techniques for Fast Machine Learning in Science
• Secure Evaluation of Quantized Neural Networks
• A Low Effort Approach to Structured CNN Design Using PCA
• Robust Adversarial Perturbation on Deep Proposal-based Models
• Analyzing Human-Human Interactions: A Survey
• Exploiting temporal and depth information for multi-frame face anti-spoofing
• ViNG: Learning Open-World Navigation with Visual Goals
• Efficient and Accurate Arbitrary-Shaped Text Detection with Pixel Aggregation Network
• Efficient Deep Learning on Multi-Source Private Data
• Optimizing CNN Model Inference on CPUs
• Pruning neural networks without any data by iteratively conserving synaptic flow
• SEED: Self-supervised Distillation For Visual Representation
• Learning N:M Fine-grained Structured Sparse Neural Networks From Scratch
• Deep Learning for Image Super-resolution: A Survey
• Improving Universal Sound Separation Using Sound Classification
• MeliusNet: Can Binary Neural Networks Achieve MobileNet-level Accuracy?
• Using Computer Vision to enhance Safety of Workforce in Manufacturing in a Post COVID World
• Learning regression and verification networks for long-term visual tracking
• UnsuperPoint: End-to-end Unsupervised Interest Point Detector and Descriptor
• LEEP: A New Measure to Evaluate Transferability of Learned Representations
• Convolutional Neural Networks with Intermediate Loss for 3D Super-Resolution of CT and MRI Scans
• PANNs: Large-Scale Pretrained Audio Neural Networks for Audio Pattern Recognition
• Dynamic Convolution: Attention over Convolution Kernels
• On the Accuracy of Analog Neural Network Inference Accelerators
• Deep Learning Framework to Detect Face Masks from Video Footage
• NSGA-Net: Neural Architecture Search using Multi-Objective Genetic Algorithm
• Sample and Computation Redistribution for Efficient Face Detection
• LaneNet: Real-Time Lane Detection Networks for Autonomous Driving
• NetAdapt: Platform-Aware Neural Network Adaptation for Mobile Applications
• Towards Practical Verification of Machine Learning: The Case of Computer Vision Systems
• Probabilistic Neural Architecture Search
• Transfer Learning-based Road Damage Detection for Multiple Countries
• OpenEDS: Open Eye Dataset
• DyNet: Dynamic Convolution for Accelerating Convolutional Neural Networks
• BNAS:An Efficient Neural Architecture Search Approach Using Broad Scalable Architecture
• Blockchain-Federated-Learning and Deep Learning Models for COVID-19 detection using CT Imaging
• How much real data do we actually need: Analyzing object detection performance using synthetic and real data
• AddNet: Deep Neural Networks Using FPGA-Optimized Multipliers
• A novel channel pruning method for deep neural network compression
• Zero-Cost Proxies for Lightweight NAS
• Axial-DeepLab: Stand-Alone Axial-Attention for Panoptic Segmentation
• A Study of Face Obfuscation in ImageNet
• Efficient Multi-objective Neural Architecture Search via Lamarckian Evolution
• EEEA-Net: An Early Exit Evolutionary Neural Architecture Search
• And the Bit Goes Down: Revisiting the Quantization of Neural Networks
• SLSNet: Skin lesion segmentation using a lightweight generative adversarial network
• ASAP: Architecture Search, Anneal and Prune
• Advanced Dropout: A Model-free Methodology for Bayesian Dropout Optimization
• Bridging the Gap Between Computational Photography and Visual Recognition
• Rethinking the Hyperparameters for Fine-tuning
• RFBNet: Deep Multimodal Networks with Residual Fusion Blocks for RGB-D Semantic Segmentation
• ResKD: Residual-Guided Knowledge Distillation
• ConTNet: Why not use convolution and transformer at the same time?
• A Survey on Deep Neural Network Compression: Challenges, Overview, and Solutions
• Discovering Neural Wirings
• PaddleSeg: A High-Efficient Development Toolkit for Image Segmentation
• On Deep Learning Techniques to Boost Monocular Depth Estimation for Autonomous Navigation
• Violence detection in videos using deep recurrent and convolutional neural networks
• Real-time Convolutional Neural Networks for Emotion and Gender Classification
• Empowering Things with Intelligence: A Survey of the Progress, Challenges, and Opportunities in Artificial Intelligence of Things
• Classification of Histopathological Biopsy Images Using Ensemble of Deep Learning Networks
• Acoustic Anomaly Detection for Machine Sounds based on Image Transfer Learning
• Cluster Pruning: An Efficient Filter Pruning Method for Edge AI Vision Applications
• Towards Better Surgical Instrument Segmentation in Endoscopic Vision: Multi-Angle Feature Aggregation and Contour Supervision
• Improving Fast Segmentation With Teacher-student Learning
• Structured Probabilistic Pruning for Convolutional Neural Network Acceleration
• On-Device Neural Net Inference with Mobile GPUs
• A Battle of Network Structures: An Empirical Study of CNN, Transformer, and MLP
• Behavioral Use Licensing for Responsible AI
• Knowledge Transfer via Distillation of Activation Boundaries Formed by Hidden Neurons
• FermiNets: Learning generative machines to generate efficient neural networks via generative synthesis
• Trained Quantization Thresholds for Accurate and Efficient Fixed-Point Inference of Deep Neural Networks
• Modeling Uncertainty with Hedged Instance Embedding
• Security Analysis of Deep Neural Networks Operating in the Presence of Cache Side-Channel Attacks
• Efficient Convolutional Neural Networks for Depth-Based Multi-Person Pose Estimation
• AlphaX: eXploring Neural Architectures with Deep Neural Networks and Monte Carlo Tree Search
• Compacting Deep Neural Networks for Internet of Things: Methods and Applications
• Driving Scene Perception Network: Real-time Joint Detection, Depth Estimation and Semantic Segmentation
• You Only Hear Once: A YOLO-like Algorithm for Audio Segmentation and Sound Event Detection
• Effective and Efficient Dropout for Deep Convolutional Neural Networks
• Lightweight Modules for Efficient Deep Learning based Image Restoration
• Q-BERT: Hessian Based Ultra Low Precision Quantization of BERT
• Restructuring Batch Normalization to Accelerate CNN Training
• Driver Behavior Recognition via Interwoven Deep Convolutional Neural Nets with Multi-stream Inputs
• ViKiNG: Vision-Based Kilometer-Scale Navigation with Geographic Hints
• MiniNet: An extremely lightweight convolutional neural network for real-time unsupervised monocular depth estimation
• Gait recognition via deep learning of the center-of-pressure trajectory
• Introduction to Camera Pose Estimation with Deep Learning
• TBC-Net: A real-time detector for infrared small target detection using semantic constraint
• A Constructive Prediction of the Generalization Error Across Scales
• Augment your batch: better training with larger batches
• pCAMP: Performance Comparison of Machine Learning Packages on the Edges
• Effects of Degradations on Deep Neural Network Architectures
• MCUNetV2: Memory-Efficient Patch-based Inference for Tiny Deep Learning
• Building Efficient ConvNets using Redundant Feature Pruning
• FedBABU: Towards Enhanced Representation for Federated Image Classification
• Learning Generalisable Omni-Scale Representations for Person Re-Identification
• MEAL V2: Boosting Vanilla ResNet-50 to 80%+ Top-1 Accuracy on ImageNet without Tricks
• Accelerator-aware Neural Network Design using AutoML
• HAQ: Hardware-Aware Automated Quantization with Mixed Precision
• Tensorized Embedding Layers for Efficient Model Compression
• Unsupervised Domain Adaptation for Mobile Semantic Segmentation based on Cycle Consistency and Feature Alignment
• Scalable Distributed DNN Training using TensorFlow and CUDA-Aware MPI: Characterization, Designs, and Performance Evaluation
• LambdaNetworks: Modeling Long-Range Interactions Without Attention
• Decision and Feature Level Fusion of Deep Features Extracted from Public COVID-19 Data-sets
• On-Device Machine Learning: An Algorithms and Learning Theory Perspective
• s-LWSR: Super Lightweight Super-Resolution Network
• MobileFaceNets: Efficient CNNs for Accurate Real-Time Face Verification on Mobile Devices
• Model Rubik's Cube: Twisting Resolution, Depth and Width for TinyNets
• Compute and memory efficient universal sound source separation
• Efficient Processing of Deep Neural Networks: A Tutorial and Survey
• Deep Mutual Learning
• ShuffleSeg: Real-time Semantic Segmentation Network
• SUNRISE: A Simple Unified Framework for Ensemble Learning in Deep Reinforcement Learning
• RepVGG: Making VGG-style ConvNets Great Again
• Reweighted Proximal Pruning for Large-Scale Language Representation
• Chart-Text: A Fully Automated Chart Image Descriptor
• BERT Loses Patience: Fast and Robust Inference with Early Exit
• K for the Price of 1: Parameter-efficient Multi-task and Transfer Learning
• Stabilizing DARTS with Amended Gradient Estimation on Architectural Parameters
• Adversarial Examples - A Complete Characterisation of the Phenomenon
• DermGAN: Synthetic Generation of Clinical Skin Images with Pathology
• DAIS: Automatic Channel Pruning via Differentiable Annealing Indicator Search
• Back to Simplicity: How to Train Accurate BNNs from Scratch?
• Depthwise Separable Convolutional ResNet with Squeeze-and-Excitation Blocks for Small-footprint Keyword Spotting
• MiniSeg: An Extremely Minimum Network for Efficient COVID-19 Segmentation
• Detecting solar system objects with convolutional neural networks
• VWA: Hardware Efficient Vectorwise Accelerator for Convolutional Neural Network
• ThunderNet: Towards Real-time Generic Object Detection
• Towards Unconstrained Palmprint Recognition on Consumer Devices: a Literature Review
• Four Things Everyone Should Know to Improve Batch Normalization
• AtomNAS: Fine-Grained End-to-End Neural Architecture Search
• Panoptic-DeepLab: A Simple, Strong, and Fast Baseline for Bottom-Up Panoptic Segmentation
• ClearBuds: Wireless Binaural Earbuds for Learning-Based Speech Enhancement
• Attention-guided Chained Context Aggregation for Semantic Segmentation
• CeyMo: See More on Roads -- A Novel Benchmark Dataset for Road Marking Detection
• Mobile Video Object Detection with Temporally-Aware Feature Maps
• Shift: A Zero FLOP, Zero Parameter Alternative to Spatial Convolutions
• Bounding Box Regression with Uncertainty for Accurate Object Detection
• Pruning Algorithms to Accelerate Convolutional Neural Networks for Edge Applications: A Survey
• Recent Advances of Continual Learning in Computer Vision: An Overview
• MobileDets: Searching for Object Detection Architectures for Mobile Accelerators
• Comparison and Benchmarking of AI Models and Frameworks on Mobile Devices
• Refiner: Refining Self-attention for Vision Transformers
• ExpandNets: Linear Over-parameterization to Train Compact Convolutional Networks
• Fingerprint Presentation Attack Detection by Channel-wise Feature Denoising
• Container: Context Aggregation Network
• YOLO-LITE: A Real-Time Object Detection Algorithm Optimized for Non-GPU Computers
• Absolute distance prediction based on deep learning object detection and monocular depth estimation models
• Rethinking Classification and Localization for Object Detection
• RWF-2000: An Open Large Scale Video Database for Violence Detection
• Efficiently utilizing complex-valued PolSAR image data via a multi-task deep learning framework
• PocketNet: A Smaller Neural Network for Medical Image Analysis
• Suppress and Balance: A Simple Gated Network for Salient Object Detection
• ASSANet: An Anisotropic Separable Set Abstraction for Efficient Point Cloud Representation Learning
• Convolution with even-sized kernels and symmetric padding
• BlockDrop: Dynamic Inference Paths in Residual Networks
• Dynamic Neural Networks: A Survey
• Improving Electron Micrograph Signal-to-Noise with an Atrous Convolutional Encoder-Decoder
• Transfer learning for radio galaxy classification
• You Only Search Once: Single Shot Neural Architecture Search via Direct Sparse Optimization
• Rethinking BiSeNet For Real-time Semantic Segmentation
• ShiftAddNet: A Hardware-Inspired Deep Network
• End-to-End Evaluation of Federated Learning and Split Learning for Internet of Things
• BigNAS: Scaling Up Neural Architecture Search with Big Single-Stage Models
• Distilling Policy Distillation
• EasyQuant: Post-training Quantization via Scale Optimization
• Multimodal and multicontrast image fusion via deep generative models
• Understanding the Disharmony between Dropout and Batch Normalization by Variance Shift
• Temporal Convolution for Real-time Keyword Spotting on Mobile Devices
• Improving Efficiency in Convolutional Neural Network with Multilinear Filters
• Generalizability vs. Robustness: Adversarial Examples for Medical Imaging
• WoodFisher: Efficient Second-Order Approximation for Neural Network Compression
• Trading-off Accuracy and Energy of Deep Inference on Embedded Systems: A Co-Design Approach
• FixyNN: Efficient Hardware for Mobile Computer Vision via Transfer Learning
• Faasm: Lightweight Isolation for Efficient Stateful Serverless Computing
• Adversarial Generation of Training Examples: Applications to Moving Vehicle License Plate Recognition
• Into the Wild with AudioScope: Unsupervised Audio-Visual Separation of On-Screen Sounds
• Exploiting Robust Unsupervised Video Person Re-identification
• Low-Memory Neural Network Training: A Technical Report
• Fast object detection in compressed JPEG Images
• Dual-stream Network for Visual Recognition
• ISTA-NAS: Efficient and Consistent Neural Architecture Search by Sparse Coding
• On-device Training: A First Overview on Existing Systems
• Contrastive Self-supervised Neural Architecture Search
• Parameter Efficient Multimodal Transformers for Video Representation Learning
• End-to-end Neural Diarization: From Transformer to Conformer
• Compounding the Performance Improvements of Assembled Techniques in a Convolutional Neural Network
• Knowledge Distillation via Route Constrained Optimization
• Medical Image Classification Using Transfer Learning and Chaos Game Optimization on the Internet of Medical Things
• EXTD: Extremely Tiny Face Detector via Iterative Filter Reuse
• Searching Efficient 3D Architectures with Sparse Point-Voxel Convolution
• When Deep Learning Meets Data Alignment: A Review on Deep Registration Networks (DRNs)
• Deformable Kernels: Adapting Effective Receptive Fields for Object Deformation
• Applying Domain Randomization to Synthetic Data for Object Category Detection
• Principled Weight Initialization for Hypernetworks
• GFD-SSD: Gated Fusion Double SSD for Multispectral Pedestrian Detection
• PFLD: A Practical Facial Landmark Detector
• Kronecker Attention Networks
• Towards High Performance Video Object Detection for Mobiles
• HAWQ: Hessian AWare Quantization of Neural Networks with Mixed-Precision
• f-CNN${}^{\text{x}}$: A Toolflow for Mapping Multi-CNN Applications on FPGAs
• Defensive Quantization: When Efficiency Meets Robustness
• A Comprehensive Survey of Machine Learning Applied to Radar Signal Processing
• Searching for Low-Bit Weights in Quantized Neural Networks
• A Comprehensive and Modularized Statistical Framework for Gradient Norm Equality in Deep Neural Networks
• Attributes Guided Feature Learning for Vehicle Re-identification
• DeepLab2: A TensorFlow Library for Deep Labeling
• Fuzzy Pooling
• Autonomous Driving with Deep Learning: A Survey of State-of-Art Technologies
• 'Skimming-Perusal' Tracking: A Framework for Real-Time and Robust Long-term Tracking
• Training Binary Neural Networks through Learning with Noisy Supervision
• Machine Learning for MU-MIMO Receive Processing in OFDM Systems
• Deep Layer Aggregation
• Reinforced Evolutionary Neural Architecture Search
• ResNet strikes back: An improved training procedure in timm
• Workpiece Image-based Tool Wear Classification in Blanking Processes Using Deep Convolutional Neural Networks
• InversionNet3D: Efficient and Scalable Learning for 3D Full Waveform Inversion
• Depthwise Convolution is All You Need for Learning Multiple Visual Domains
• StressNAS: Affect State and Stress Detection Using Neural Architecture Search
• TinyML for Ubiquitous Edge AI
• TinyTL: Reduce Activations, Not Trainable Parameters for Efficient On-Device Learning
• Edge AI: On-Demand Accelerating Deep Neural Network Inference via Edge Computing
• torchgpipe: On-the-fly Pipeline Parallelism for Training Giant Models
• An Analysis of SVD for Deep Rotation Estimation
• Stacked U-Nets: A No-Frills Approach to Natural Image Segmentation
• A Non-Technical Survey on Deep Convolutional Neural Network Architectures
• Towards Optimal Structured CNN Pruning via Generative Adversarial Learning
• Understanding Knowledge Distillation in Non-autoregressive Machine Translation
• Smartphone Sensing for the Well-being of Young Adults: A Review
• Mind the Pad -- CNNs can Develop Blind Spots
• DSConv: Efficient Convolution Operator
• Accurate and Compact Convolutional Neural Networks with Trained Binarization
• TAda! Temporally-Adaptive Convolutions for Video Understanding
• Training Competitive Binary Neural Networks from Scratch
• QuartzNet: Deep Automatic Speech Recognition with 1D Time-Channel Separable Convolutions
• ROMANet: Fine-Grained Reuse-Driven Off-Chip Memory Access Management and Data Organization for Deep Neural Network Accelerators
• Feature Fusion for Online Mutual Knowledge Distillation
• BinaryDuo: Reducing Gradient Mismatch in Binary Activation Network by Coupling Binary Activations
• OmniPose: A Multi-Scale Framework for Multi-Person Pose Estimation
• Defending Against Machine Learning Model Stealing Attacks Using Deceptive Perturbations
• Zero Time Waste: Recycling Predictions in Early Exit Neural Networks
• EdgeSpeechNets: Highly Efficient Deep Neural Networks for Speech Recognition on the Edge
• Looking Fast and Slow: Memory-Guided Mobile Video Object Detection
• DenseBody: Directly Regressing Dense 3D Human Pose and Shape From a Single Color Image
• Fixed-Point Convolutional Neural Network for Real-Time Video Processing in FPGA
• AM-MobileNet1D: A Portable Model for Speaker Recognition
• SqueezeSegV3: Spatially-Adaptive Convolution for Efficient Point-Cloud Segmentation
• GOLD-NAS: Gradual, One-Level, Differentiable
• Is it enough to optimize CNN architectures on ImageNet?
• S${}^2$-MLP: Spatial-Shift MLP Architecture for Vision
• Data-Driven Sparse Structure Selection for Deep Neural Networks
• FBNetV2: Differentiable Neural Architecture Search for Spatial and Channel Dimensions
• Glance and Focus: a Dynamic Approach to Reducing Spatial Redundancy in Image Classification
• Musical Chair: Efficient Real-Time Recognition Using Collaborative IoT Devices
• Fast, nonlocal and neural: a lightweight high quality solution to image denoising
• ReActNet: Towards Precise Binary Neural Network with Generalized Activation Functions
• Dynamic Spatio-temporal Graph-based CNNs for Traffic Prediction
• FEELVOS: Fast End-to-End Embedding Learning for Video Object Segmentation
• A Survey of Modern Object Detection Literature using Deep Learning
• Reveal of Domain Effect: How Visual Restoration Contributes to Object Detection in Aquatic Scenes
• Weight-Sharing Neural Architecture Search: A Battle to Shrink the Optimization Gap
• FedBE: Making Bayesian Model Ensemble Applicable to Federated Learning
• Using Videos to Evaluate Image Model Robustness
• Scaling Wide Residual Networks for Panoptic Segmentation
• WIDER Face and Pedestrian Challenge 2018: Methods and Results
• RefineDetLite: A Lightweight One-stage Object Detection Framework for CPU-only Devices
• Large-Scale Generative Data-Free Distillation
• Learning Sparse Low-Precision Neural Networks With Learnable Regularization
• CVR-Net: A deep convolutional neural network for coronavirus recognition from chest radiography images
• COVID-19 Screening Using Residual Attention Network an Artificial Intelligence Approach
• Dynamic Sparse Training: Find Efficient Sparse Network From Scratch With Trainable Masked Layers
• FabricNet: A Fiber Recognition Architecture Using Ensemble ConvNets
• Memory-Driven Mixed Low Precision Quantization For Enabling Deep Network Inference On Microcontrollers
• IGCV3: Interleaved Low-Rank Group Convolutions for Efficient Deep Neural Networks
• Deep Association Learning for Unsupervised Video Person Re-identification
• Channel Compression: Rethinking Information Redundancy among Channels in CNN Architecture
• MobilePose: Real-Time Pose Estimation for Unseen Objects with Weak Shape Supervision
• Light-Weight RetinaNet for Object Detection
• Improving Object Detection from Scratch via Gated Feature Reuse
• RNNPool: Efficient Non-linear Pooling for RAM Constrained Inference
• RESA: Recurrent Feature-Shift Aggregator for Lane Detection
• ShrinkTeaNet: Million-scale Lightweight Face Recognition via Shrinking Teacher-Student Networks
• Scaling Laws for Deep Learning
• DSNAS: Direct Neural Architecture Search without Parameter Retraining
• How Does Batch Normalization Help Binary Training?
• Learning Guided Convolutional Network for Depth Completion
• Pareto-Optimal Quantized ResNet Is Mostly 4-bit
• VID-WIN: Fast Video Event Matching with Query-Aware Windowing at the Edge for the Internet of Multimedia Things
• Multi-frame Feature Aggregation for Real-time Instrument Segmentation in Endoscopic Video
• Latency-Aware Differentiable Neural Architecture Search
• Scaling Video Analytics on Constrained Edge Nodes
• IamNN: Iterative and Adaptive Mobile Neural Network for Efficient Image Classification
• Preferences Prediction using a Gallery of Mobile Device based on Scene Recognition and Object Detection
• PP-OCRv2: Bag of Tricks for Ultra Lightweight OCR System
• AI Benchmark: Running Deep Neural Networks on Android Smartphones
• Training Shallow and Thin Networks for Acceleration via Knowledge Distillation with Conditional Adversarial Networks
• Diagnosis of Autism in Children using Facial Analysis and Deep Learning
• RC-DARTS: Resource Constrained Differentiable Architecture Search
• FedGEMS: Federated Learning of Larger Server Models via Selective Knowledge Fusion
• ConvMLP: Hierarchical Convolutional MLPs for Vision
• CEREALS - Cost-Effective REgion-based Active Learning for Semantic Segmentation
• Towards Real-World Blind Face Restoration with Generative Facial Prior
• iQIYI-VID: A Large Dataset for Multi-modal Person Identification
• Recent Advances in Deep Learning for Object Detection
• MARS: Multi-macro Architecture SRAM CIM-Based Accelerator with Co-designed Compressed Neural Networks
• AGDC: Automatic Garbage Detection and Collection
• Prototype Guided Federated Learning of Visual Feature Representations
• Nimble: Efficiently Compiling Dynamic Neural Networks for Model Inference
• Computing Systems for Autonomous Driving: State-of-the-Art and Challenges
• CSL-YOLO: A New Lightweight Object Detection System for Edge Computing
• Task dependent Deep LDA pruning of neural networks
• Theory-Inspired Path-Regularized Differential Network Architecture Search
• A Performance Comparison of Loss Functions for Deep Face Recognition
• VarGNet: Variable Group Convolutional Neural Network for Efficient Embedded Computing
• Joint Multi-Dimension Pruning via Numerical Gradient Update
• RP2K: A Large-Scale Retail Product Dataset for Fine-Grained Image Classification
• AclNet: efficient end-to-end audio classification CNN
• Quantization for Rapid Deployment of Deep Neural Networks
• DeepReDuce: ReLU Reduction for Fast Private Inference
• PAN++: Towards Efficient and Accurate End-to-End Spotting of Arbitrarily-Shaped Text
• NeurObfuscator: A Full-stack Obfuscation Tool to Mitigate Neural Architecture Stealing
• WebFace260M: A Benchmark Unveiling the Power of Million-Scale Deep Face Recognition
• Auto-Icon+: An Automated End-to-End Code Generation Tool for Icon Designs in UI Development
• Identification of Pediatric Respiratory Diseases Using Fine-grained Diagnosis System
• SquishedNets: Squishing SqueezeNet further for edge device scenarios via deep evolutionary synthesis
• PCONV: The Missing but Desirable Sparsity in DNN Weight Pruning for Real-time Execution on Mobile Devices
• Similarity of Neural Network Models: A Survey of Functional and Representational Measures
• Federated Neural Architecture Search
• A Classification Supervised Auto-Encoder Based on Predefined Evenly-Distributed Class Centroids
• Duckiefloat: a Collision-Tolerant Resource-Constrained Blimp for Long-Term Autonomy in Subterranean Environments
• Addressing Missing Labels in Large-Scale Sound Event Recognition Using a Teacher-Student Framework With Loss Masking
• Constructing Fast Network through Deconstruction of Convolution
• An efficient solution for semantic segmentation: ShuffleNet V2 with atrous separable convolutions
• Automatic Identification of MHD Modes in Magnetic Fluctuations Spectrograms using Deep Learning Techniques
• DiPair: Fast and Accurate Distillation for Trillion-Scale Text Matching and Pair Modeling
• Image Classification with Classic and Deep Learning Techniques
• Self-Binarizing Networks
• Towards Robust Learning-Based Pose Estimation of Noncooperative Spacecraft
• End-to-End Supermask Pruning: Learning to Prune Image Captioning Models
• Diverse Branch Block: Building a Convolution as an Inception-like Unit
• Packing Sparse Convolutional Neural Networks for Efficient Systolic Array Implementations: Column Combining Under Joint Optimization
• AdderNet and its Minimalist Hardware Design for Energy-Efficient Artificial Intelligence
• Sequential Image-based Attention Network for Inferring Force Estimation without Haptic Sensor
• Local-to-Global Self-Attention in Vision Transformers
• What does it mean to understand a neural network?
• Towards Effective Low-bitwidth Convolutional Neural Networks
• Dataset Distillation
• Towards Palmprint Verification On Smartphones
• Vehicle Re-Identification: an Efficient Baseline Using Triplet Embedding
• FBNetV3: Joint Architecture-Recipe Search using Predictor Pretraining
• Identity-Driven DeepFake Detection
• APQ: Joint Search for Network Architecture, Pruning and Quantization Policy
• AdaSpring: Context-adaptive and Runtime-evolutionary Deep Model Compression for Mobile Applications
• Characterizing signal propagation to close the performance gap in unnormalized ResNets
• LEDNet: A Lightweight Encoder-Decoder Network for Real-Time Semantic Segmentation
• Randomness In Neural Network Training: Characterizing The Impact of Tooling
• GhostSR: Learning Ghost Features for Efficient Image Super-Resolution
• Attention-based Dropout Layer for Weakly Supervised Object Localization
• Wisdom of Committees: An Overlooked Approach To Faster and More Accurate Models
• Probabilistic Dual Network Architecture Search on Graphs
• Model Slicing for Supporting Complex Analytics with Elastic Inference Cost and Resource Constraints
• Accelerate CNNs from Three Dimensions: A Comprehensive Pruning Framework
• Lightweight Residual Densely Connected Convolutional Neural Network
• SqueezeWave: Extremely Lightweight Vocoders for On-device Speech Synthesis
• Advanced Capsule Networks via Context Awareness
• Utilizing Smartphone-Based Machine Learning in Medical Monitor Data Collection: Seven Segment Digit Recognition
• Quantizing Convolutional Neural Networks for Low-Power High-Throughput Inference Engines
• MAMNet: Multi-path Adaptive Modulation Network for Image Super-Resolution
• Efficient, high-performance pancreatic segmentation using multi-scale feature extraction
• YOLoC: DeploY Large-Scale Neural Network by ROM-based Computing-in-Memory using ResiduaL Branch on a Chip
• Private Model Compression via Knowledge Distillation
• Dog Identification using Soft Biometrics and Neural Networks
• Not All Ops Are Created Equal!
• Fighting Quantization Bias With Bias
• Characterizing the Deep Neural Networks Inference Performance of Mobile Applications
• SkyNet: A Champion Model for DAC-SDC on Low Power Object Detection
• Mitigating Edge Machine Learning Inference Bottlenecks: An Empirical Study on Accelerating Google Edge Models
• Resource-Efficient Neural Networks for Embedded Systems
• Progressive DARTS: Bridging the Optimization Gap for NAS in the Wild
• STEERAGE: Synthesis of Neural Networks Using Architecture Search and Grow-and-Prune Methods
• Improved Techniques for Training Adaptive Deep Networks
• Sparse Transfer Learning via Winning Lottery Tickets
• Towards Practical 2D Grapevine Bud Detection with Fully Convolutional Networks
• MicroNet: Towards Image Recognition with Extremely Low FLOPs
• IPGuard: Protecting Intellectual Property of Deep Neural Networks via Fingerprinting the Classification Boundary
• Rethinking Text Line Recognition Models
• Industrial Scale Privacy Preserving Deep Neural Network
• Rate Distortion For Model Compression: From Theory To Practice
• ZeroQ: A Novel Zero Shot Quantization Framework
• Differentiable Model Compression via Pseudo Quantization Noise
• Splitting Steepest Descent for Growing Neural Architectures
• Not All Images are Worth 16x16 Words: Dynamic Transformers for Efficient Image Recognition
• Joint Neural Architecture Search and Quantization
• Accuracy-Efficiency Trade-Offs and Accountability in Distributed ML Systems
• Deep Learning for Semantic Segmentation on Minimal Hardware
• ExtremeC3Net: Extreme Lightweight Portrait Segmentation Networks using Advanced C3-modules
• Feature Pyramid Encoding Network for Real-time Semantic Segmentation
• Marvel: A Data-centric Compiler for DNN Operators on Spatial Accelerators
• Feature Distillation: DNN-Oriented JPEG Compression Against Adversarial Examples
• CityFlow: A City-Scale Benchmark for Multi-Target Multi-Camera Vehicle Tracking and Re-Identification
• Personalized Federated Deep Learning for Pain Estimation From Face Images
• FoodTracker: A Real-time Food Detection Mobile Application by Deep Convolutional Neural Networks
• ANTNets: Mobile Convolutional Neural Networks for Resource Efficient Image Classification
• DPP-Net: Device-aware Progressive Search for Pareto-optimal Neural Architectures
• Efficient Accelerator for Dilated and Transposed Convolution with Decomposition
• Optimizing speed/accuracy trade-off for person re-identification via knowledge distillation
• Weight-dependent Gates for Network Pruning
• A Dataset and Benchmark Towards Multi-Modal Face Anti-Spoofing Under Surveillance Scenarios
• A Panda? No, It's a Sloth: Slowdown Attacks on Adaptive Multi-Exit Neural Network Inference
• Deep4Air: A Novel Deep Learning Framework for Airport Airside Surveillance
• Efficient Visual Recognition with Deep Neural Networks: A Survey on Recent Advances and New Directions
• FRILL: A Non-Semantic Speech Embedding for Mobile Devices
• MANAS: Multi-Agent Neural Architecture Search
• A Delay Metric for Video Object Detection: What Average Precision Fails to Tell
• Towards Privacy-Preserving Visual Recognition via Adversarial Training: A Pilot Study
• Dynamic Instance Normalization for Arbitrary Style Transfer
• Finite size corrections for neural network Gaussian processes
• Deep Learning based approach to detect Customer Age, Gender and Expression in Surveillance Video
• Projection Convolutional Neural Networks for 1-bit CNNs via Discrete Back Propagation
• Mixed-precision Neural Networks on RISC-V Cores: ISA extensions for Multi-Pumped Soft SIMD Operations
• Comparison of State-of-the-Art Deep Learning APIs for Image Multi-Label Classification using Semantic Metrics
• ChamNet: Towards Efficient Network Design through Platform-Aware Model Adaptation
• ReNAS:Relativistic Evaluation of Neural Architecture Search
• Optimizing the Trade-off between Single-Stage and Two-Stage Object Detectors using Image Difficulty Prediction
• Auto-Panoptic: Cooperative Multi-Component Architecture Search for Panoptic Segmentation
• Design Automation for Efficient Deep Learning Computing
• Metamorphic Testing for Object Detection Systems
• End-to-End Entity Classification on Multimodal Knowledge Graphs
• Learned Low Precision Graph Neural Networks
• Adversarial Infidelity Learning for Model Interpretation
• Universally Slimmable Networks and Improved Training Techniques
• AdaBits: Neural Network Quantization with Adaptive Bit-Widths
• MultiScene: A Large-scale Dataset and Benchmark for Multi-scene Recognition in Single Aerial Images
• Chameleon: Adaptive Code Optimization for Expedited Deep Neural Network Compilation
• HardCoRe-NAS: Hard Constrained diffeRentiable Neural Architecture Search
• Siamese Box Adaptive Network for Visual Tracking
• Switchable Precision Neural Networks
• Real-time Federated Evolutionary Neural Architecture Search
• Soft-Root-Sign Activation Function
• Breaking Batch Normalization for better explainability of Deep Neural Networks through Layer-wise Relevance Propagation
• Cross-dataset Training for Class Increasing Object Detection
• FQ-Conv: Fully Quantized Convolution for Efficient and Accurate Inference
• Hardware-Centric AutoML for Mixed-Precision Quantization
• Imperceptible Adversarial Examples by Spatial Chroma-Shift
• CFPNet-M: A Light-Weight Encoder-Decoder Based Network for Multimodal Biomedical Image Real-Time Segmentation
• Intersection focused Situation Coverage-based Verification and Validation Framework for Autonomous Vehicles Implemented in CARLA
• Pruning Self-attentions into Convolutional Layers in Single Path
• DSPoint: Dual-scale Point Cloud Recognition with High-frequency Fusion
• Fast and Flexible Human Pose Estimation with HyperPose
• I-BERT: Integer-only BERT Quantization
• How Much Can We Really Trust You? Towards Simple, Interpretable Trust Quantification Metrics for Deep Neural Networks
• Post-Training Piecewise Linear Quantization for Deep Neural Networks
• Knowledge distillation via adaptive instance normalization
• Adapted Center and Scale Prediction: More Stable and More Accurate
• Percival: Making In-Browser Perceptual Ad Blocking Practical With Deep Learning
• Structured Knowledge Distillation for Dense Prediction
• Kernel Transformer Networks for Compact Spherical Convolution
• Graph-Based Global Reasoning Networks
• Structured Binary Neural Networks for Accurate Image Classification and Semantic Segmentation
• Relaxed Quantization for Discretized Neural Networks
• Learning to Train a Binary Neural Network
• Segmentation of Liver Lesions with Reduced Complexity Deep Models
• Target Driven Instance Detection
• The iNaturalist Species Classification and Detection Dataset
• Skin disease identification from dermoscopy images using deep convolutional neural network
• Pooling Pyramid Network for Object Detection
• Deep Learning Assessment of galaxy morphology in S-PLUS DataRelease 1
• AdaViT: Adaptive Vision Transformers for Efficient Image Recognition
• Taxonomy and Evaluation of Structured Compression of Convolutional Neural Networks
• Addressing the Cold-Start Problem in Outfit Recommendation Using Visual Preference Modelling
• EdgeNet: Balancing Accuracy and Performance for Edge-based Convolutional Neural Network Object Detectors
• KTAN: Knowledge Transfer Adversarial Network
• A deep learning based solution for construction equipment detection: from development to deployment
• Transfer Learning for Oral Cancer Detection using Microscopic Images
• Hierarchical Dynamic Filtering Network for RGB-D Salient Object Detection
• Improving Prostate Cancer Detection with Breast Histopathology Images
• Privacy Aware Offloading of Deep Neural Networks
• Understanding Reuse, Performance, and Hardware Cost of DNN Dataflows: A Data-Centric Approach Using MAESTRO
• Continual 3D Convolutional Neural Networks for Real-time Processing of Videos
• Efficient Neural Network Training via Forward and Backward Propagation Sparsification
• MoBiNet: A Mobile Binary Network for Image Classification
• Multinomial Distribution Learning for Effective Neural Architecture Search
• Funnel Activation for Visual Recognition
• Rafiki: Machine Learning as an Analytics Service System
• LeanConvNets: Low-cost Yet Effective Convolutional Neural Networks
• SuPer Deep: A Surgical Perception Framework for Robotic Tissue Manipulation using Deep Learning for Feature Extraction
• Exploiting Errors for Efficiency: A Survey from Circuits to Algorithms
• PyHealth: A Python Library for Health Predictive Models
• Exchangeable deep neural networks for set-to-set matching and learning
• Evolutionary-Neural Hybrid Agents for Architecture Search
• Single-Label Multi-Class Image Classification by Deep Logistic Regression
• Multi-Objective Automatic Machine Learning with AutoxgboostMC
• Biased Mixtures Of Experts: Enabling Computer Vision Inference Under Data Transfer Limitations
• Deep Mangoes: from fruit detection to cultivar identification in colour images of mango trees
• C3: Concentrated-Comprehensive Convolution and its application to semantic segmentation
• Efficient Object Detection Model for Real-Time UAV Applications
• An Experimental Study of the Impact of Pre-training on the Pruning of a Convolutional Neural Network
• Pixel Difference Networks for Efficient Edge Detection
• DarKnight: A Data Privacy Scheme for Training and Inference of Deep Neural Networks
• MEAL: Multi-Model Ensemble via Adversarial Learning
• MirBot: A collaborative object recognition system for smartphones using convolutional neural networks
• ASFD: Automatic and Scalable Face Detector
• Densely Guided Knowledge Distillation using Multiple Teacher Assistants
• Tracking-by-Trackers with a Distilled and Reinforced Model
• Rethinking Channel Dimensions for Efficient Model Design
• ZigZag: A Memory-Centric Rapid DNN Accelerator Design Space Exploration Framework
• Searching for Winograd-aware Quantized Networks
• AReLU: Attention-based Rectified Linear Unit
• SCAN: A Scalable Neural Networks Framework Towards Compact and Efficient Models
• CondenseNeXt: An Ultra-Efficient Deep Neural Network for Embedded Systems
• AdderNet: Do We Really Need Multiplications in Deep Learning?
• Mobile Video Action Recognition
• EffNet: An Efficient Structure for Convolutional Neural Networks
• Learning Efficient Detector with Semi-supervised Adaptive Distillation
• HPLFlowNet: Hierarchical Permutohedral Lattice FlowNet for Scene Flow Estimation on Large-scale Point Clouds
• STONNE: A Detailed Architectural Simulator for Flexible Neural Network Accelerators
• Generalization Bounds for Convolutional Neural Networks
• Dynamic Mini-batch SGD for Elastic Distributed Training: Learning in the Limbo of Resources
• Efficient Differentiable Neural Architecture Search with Meta Kernels
• Zero-Shot Knowledge Distillation from a Decision-Based Black-Box Model
• To Ensemble or Not Ensemble: When does End-To-End Training Fail?
• DeepObfuscation: Securing the Structure of Convolutional Neural Networks via Knowledge Distillation
• The Architectural Implications of Facebook's DNN-based Personalized Recommendation
• Layer Pruning via Fusible Residual Convolutional Block for Deep Neural Networks
• FD-MobileNet: Improved MobileNet with a Fast Downsampling Strategy
• DeepAtom: A Framework for Protein-Ligand Binding Affinity Prediction
• AirSim Drone Racing Lab
• One-Shot Pruning of Recurrent Neural Networks by Jacobian Spectrum Evaluation
• Rethinking Bottleneck Structure for Efficient Mobile Network Design
• Efficient 3D Fully Convolutional Networks for Pulmonary Lobe Segmentation in CT Images
• A CNN-Based Blind Denoising Method for Endoscopic Images
• MobileStyleGAN: A Lightweight Convolutional Neural Network for High-Fidelity Image Synthesis
• WQT and DG-YOLO: towards domain generalization in underwater object detection
• Multi-Fiber Networks for Video Recognition
• Advanced Deep Learning Methodologies for Skin Cancer Classification in Prodromal Stages
• Reinforcement Learning and Adaptive Sampling for Optimized DNN Compilation
• EagleEye: Fast Sub-net Evaluation for Efficient Neural Network Pruning
• Neural networks on microcontrollers: saving memory at inference via operator reordering
• Group Ensemble: Learning an Ensemble of ConvNets in a single ConvNet
• CoCoPIE: Making Mobile AI Sweet As PIE --Compression-Compilation Co-Design Goes a Long Way
• Comparing Computing Platforms for Deep Learning on a Humanoid Robot
• Dense Crowds Detection and Counting with a Lightweight Architecture
• Shift-based Primitives for Efficient Convolutional Neural Networks
• Learning to play the Chess Variant Crazyhouse above World Champion Level with Deep Neural Networks and Human Data
• Fine-Grained Neural Architecture Search
• Efficient Segmentation: Learning Downsampling Near Semantic Boundaries
• OpenEI: An Open Framework for Edge Intelligence
• Restricted Recurrent Neural Networks
• Optimal checkpointing for heterogeneous chains: how to train deep neural networks with limited memory
• A White Paper on Neural Network Quantization
• CNN-based Facial Affect Analysis on Mobile Devices
• Orchestrating the Development Lifecycle of Machine Learning-Based IoT Applications: A Taxonomy and Survey
• Improving Device-Edge Cooperative Inference of Deep Learning via 2-Step Pruning
• Learnable Embedding Space for Efficient Neural Architecture Compression
• Densely Connected Search Space for More Flexible Neural Architecture Search
• BlockSwap: Fisher-guided Block Substitution for Network Compression on a Budget
• Byzantine-Robust and Privacy-Preserving Framework for FedML
• Continual Learning on the Edge with TensorFlow Lite
• Knowledge Squeezed Adversarial Network Compression
• Pruning artificial neural networks: a way to find well-generalizing, high-entropy sharp minima
• Highly Efficient Salient Object Detection with 100K Parameters
• Fix your classifier: the marginal value of training the last weight layer
• RGBD Based Dimensional Decomposition Residual Network for 3D Semantic Scene Completion
• Adaptive Fractional Dilated Convolution Network for Image Aesthetics Assessment
• Optimization of Quantum-dot Qubit Fabrication via Machine Learning
• Indices Matter: Learning to Index for Deep Image Matting
• LF-YOLO: A Lighter and Faster YOLO for Weld Defect Detection of X-ray Image
• Generalization bounds for deep learning
• A Comprehensive Overhaul of Feature Distillation
• Road Segmentation Using CNN with GRU
• Feature Pyramid Grids
• Real-Time High-Performance Semantic Image Segmentation of Urban Street Scenes
• RotDCF: Decomposition of Convolutional Filters for Rotation-Equivariant Deep Networks
• Pose Neural Fabrics Search
• LIT: Block-wise Intermediate Representation Training for Model Compression
• A Framework of Transfer Learning in Object Detection for Embedded Systems
• SGAS: Sequential Greedy Architecture Search
• Revisiting Dynamic Convolution via Matrix Decomposition
• DAC-SDC Low Power Object Detection Challenge for UAV Applications
• An Embarrassingly Simple Approach for Knowledge Distillation
• Deep Multi-Resolution Dictionary Learning for Histopathology Image Analysis
• BoolNet: Minimizing The Energy Consumption of Binary Neural Networks
• SIGAN: A Novel Image Generation Method for Solar Cell Defect Segmentation and Augmentation
• DHP: Differentiable Meta Pruning via HyperNetworks
• SSFL: Tackling Label Deficiency in Federated Learning via Personalized Self-Supervision
• Deep Reasoning with Multi-Scale Context for Salient Object Detection
• AttentiveNAS: Improving Neural Architecture Search via Attentive Sampling
• Multi-Scale RCNN Model for Financial Time-series Classification
• Control Distance IoU and Control Distance IoU Loss Function for Better Bounding Box Regression
• MicroRec: Efficient Recommendation Inference by Hardware and Data Structure Solutions
• Towards Compact ConvNets via Structure-Sparsity Regularized Filter Pruning
• Long Short-Term Transformer for Online Action Detection
• Distilling Knowledge via Knowledge Review
• C3AE: Exploring the Limits of Compact Model for Age Estimation
• Neural Architecture Design for GPU-Efficient Networks
• An Information Theory-inspired Strategy for Automatic Network Pruning
• Equivalent and Approximate Transformations of Deep Neural Networks
• Lite-HRNet: A Lightweight High-Resolution Network
• LAFFNet: A Lightweight Adaptive Feature Fusion Network for Underwater Image Enhancement
• EdgeCNN: Convolutional Neural Network Classification Model with small inputs for Edge Computing
• Towards Creating a Deployable Grasp Type Probability Estimator for a Prosthetic Hand
• Optimizing Video Object Detection via a Scale-Time Lattice
• Building Computationally Efficient and Well-Generalizing Person Re-Identification Models with Metric Learning
• An Analysis of Deep Object Detectors For Diver Detection
• Automated Design Space Exploration for optimised Deployment of DNN on Arm Cortex-A CPUs
• Generalization through Simulation: Integrating Simulated and Real Data into Deep Reinforcement Learning for Vision-Based Autonomous Flight
• ACTION-Net: Multipath Excitation for Action Recognition
• All You Need is a Few Shifts: Designing Efficient Convolutional Neural Networks for Image Classification
• Rethinking Differentiable Search for Mixed-Precision Neural Networks
• Faa$T: A Transparent Auto-Scaling Cache for Serverless Applications
• Enabling Large Neural Networks on Tiny Microcontrollers with Swapping
• GeneCAI: Genetic Evolution for Acquiring Compact AI
• Differentiable Architecture Search with Ensemble Gumbel-Softmax
• AC/DC: Alternating Compressed/DeCompressed Training of Deep Neural Networks
• Differentiable Fine-grained Quantization for Deep Neural Network Compression
• Constrained Deep Learning using Conditional Gradient and Applications in Computer Vision
• Eyeriss v2: A Flexible Accelerator for Emerging Deep Neural Networks on Mobile Devices
• Ternary Hybrid Neural-Tree Networks for Highly Constrained IoT Applications
• Understanding Generalization in Deep Learning via Tensor Methods
• Adversarial Alignment of Class Prediction Uncertainties for Domain Adaptation
• BLK-REW: A Unified Block-based DNN Pruning Framework using Reweighted Regularization Method
• Explicit Shape Encoding for Real-Time Instance Segmentation
• Weight Normalization based Quantization for Deep Neural Network Compression
• PIRM Challenge on Perceptual Image Enhancement on Smartphones: Report
• Highly Efficient Natural Image Matting
• Rethinking FUN: Frequency-Domain Utilization Networks
• ApproxNet: Content and Contention-Aware Video Analytics System for Embedded Clients
• Ensemble Knowledge Distillation for Learning Improved and Efficient Networks
• Cyclic Differentiable Architecture Search
• NASGEM: Neural Architecture Search via Graph Embedding Method
• How Do the Hearts of Deep Fakes Beat? Deep Fake Source Detection via Interpreting Residuals with Biological Signals
• Learning Deep Representations with Probabilistic Knowledge Transfer
• Deep Face Recognition Model Compression via Knowledge Transfer and Distillation
• A Survey on Deep Learning Methods for Semantic Image Segmentation in Real-Time
• Activate or Not: Learning Customized Activation
• Efficient Deep Neural Networks
• Deep Density-aware Count Regressor
• Neural Architecture Search using Deep Neural Networks and Monte Carlo Tree Search
• Fingerprint Spoof Generalization
• Real-time Denoising and Dereverberation with Tiny Recurrent U-Net
• Hard-Attention for Scalable Image Classification
• AI Blue Book: Vehicle Price Prediction using Visual Features
• Ego-Pose Estimation and Forecasting as Real-Time PD Control
• Fast and Efficient Zero-Learning Image Fusion
• Explaining in Style: Training a GAN to explain a classifier in StyleSpace
• Efficient Neural Architecture Search via Proximal Iterations
• Joint Deep Cross-Domain Transfer Learning for Emotion Recognition
• AR-Net: Adaptive Frame Resolution for Efficient Action Recognition
• XSepConv: Extremely Separated Convolution
• Neural Architecture Search for Deep Face Recognition
• MSG-Transformer: Exchanging Local Spatial Information by Manipulating Messenger Tokens
• Sanity Checks for Lottery Tickets: Does Your Winning Ticket Really Win the Jackpot?
• DeepLight: Robust & Unobtrusive Real-time Screen-Camera Communication for Real-World Displays
• Exploring the Applications of Faster R-CNN and Single-Shot Multi-box Detection in a Smart Nursery Domain
• Traceability of Deep Neural Networks
• Embedded Real-Time Fall Detection Using Deep Learning For Elderly Care
• A3D: Adaptive 3D Networks for Video Action Recognition
• Evolutionary Neural AutoML for Deep Learning
• Minimum weight norm models do not always generalize well for over-parameterized problems
• SEFR: A Fast Linear-Time Classifier for Ultra-Low Power Devices
• Face Detection with Feature Pyramids and Landmarks
• A Distributed Hierarchical SGD Algorithm with Sparse Global Reduction
• Deep Learning in Diabetic Foot Ulcers Detection: A Comprehensive Evaluation
• Efficient Inference of CNNs via Channel Pruning
• Handwriting recognition and automatic scoring for descriptive answers in Japanese language tests
• Compact Global Descriptor for Neural Networks
• Depthwise Multiception Convolution for Reducing Network Parameters without Sacrificing Accuracy
• Balanced One-shot Neural Architecture Optimization
• Training Compact Neural Networks with Binary Weights and Low Precision Activations
• Dynamic Region-Aware Convolution
• Assurance Monitoring of Cyber-Physical Systems with Machine Learning Components
• MSCFNet: A Lightweight Network With Multi-Scale Context Fusion for Real-Time Semantic Segmentation
• Rethinking Token-Mixing MLP for MLP-based Vision Backbone
• Fingerprints: Fixed Length Representation via Deep Networks and Domain Knowledge
• Deep Learning Training on the Edge with Low-Precision Posits
• FASTER Recurrent Networks for Efficient Video Classification
• LeYOLO, New Embedded Architecture for Object Detection
• DeepWaste: Applying Deep Learning to Waste Classification for a Sustainable Planet
• GazeGAN - Unpaired Adversarial Image Generation for Gaze Estimation
• Intra-model Variability in COVID-19 Classification Using Chest X-ray Images
• Building an Integrated Mobile Robotic System for Real-Time Applications in Construction
• Rethinking Machine Learning Development and Deployment for Edge Devices
• Deepfake Video Forensics based on Transfer Learning
• Growing Efficient Deep Networks by Structured Continuous Sparsification
• HCMS: Hierarchical and Conditional Modality Selection for Efficient Video Recognition
• MnasFPN: Learning Latency-aware Pyramid Architecture for Object Detection on Mobile Devices
• Effective Training of Convolutional Neural Networks with Low-bitwidth Weights and Activations
• SwishNet: A Fast Convolutional Neural Network for Speech, Music and Noise Classification and Segmentation
• DNNVM : End-to-End Compiler Leveraging Heterogeneous Optimizations on FPGA-based CNN Accelerators
• Partial Order Pruning: for Best Speed/Accuracy Trade-off in Neural Architecture Search
• Wildfire Smoke Detection System: Model Architecture, Training Mechanism, and Dataset
• Energy-Aware Neural Architecture Optimization with Fast Splitting Steepest Descent
• NASI: Label- and Data-agnostic Neural Architecture Search at Initialization
• AnalogNets: ML-HW Co-Design of Noise-robust TinyML Models and Always-On Analog Compute-in-Memory Accelerator
• Anycost GANs for Interactive Image Synthesis and Editing
• LUTNet: Rethinking Inference in FPGA Soft Logic
• OMPQ: Orthogonal Mixed Precision Quantization
• Structured Deep Neural Network Pruning via Matrix Pivoting
• Benanza: Automatic $\mu$Benchmark Generation to Compute "Lower-bound" Latency and Inform Optimizations of Deep Learning Models on GPUs
• CodeX: Bit-Flexible Encoding for Streaming-based FPGA Acceleration of DNNs
• TF-NAS: Rethinking Three Search Freedoms of Latency-Constrained Differentiable Neural Architecture Search
• Organ Segmentation From Full-size CT Images Using Memory-Efficient FCN
• Towards Improving the Consistency, Efficiency, and Flexibility of Differentiable Neural Architecture Search
• Wise-SrNet: A Novel Architecture for Enhancing Image Classification by Learning Spatial Resolution of Feature Maps
• Depth-wise Decomposition for Accelerating Separable Convolutions in Efficient Convolutional Neural Networks
• Low-Complexity Models for Acoustic Scene Classification Based on Receptive Field Regularization and Frequency Damping
• Who wants accurate models? Arguing for a different metrics to take classification models seriously
• SwiftNet: Using Graph Propagation as Meta-knowledge to Search Highly Representative Neural Architectures
• PhotoSafer: Content-Based and Context-Aware Private Photo Protection for Smartphones
• LiteDenseNet: A Lightweight Network for Hyperspectral Image Classification
• Feature Pyramid Network for Multi-task Affective Analysis
• Fully Point-wise Convolutional Neural Network for Modeling Statistical Regularities in Natural Images
• Bi-Real Net: Binarizing Deep Network Towards Real-Network Performance
• SmartExchange: Trading Higher-cost Memory Storage/Access for Lower-cost Computation
• AirFace: Lightweight and Efficient Model for Face Recognition
• Assessing a mobile-based deep learning model for plant disease surveillance
• Exploration of Quantum Neural Architecture by Mixing Quantum Neuron Designs
• Automated Model Design and Benchmarking of 3D Deep Learning Models for COVID-19 Detection with Chest CT Scans
• S2DNAS:Transforming Static CNN Model for Dynamic Inference via Neural Architecture Search
• FSNet: Compression of Deep Convolutional Neural Networks by Filter Summary
• Learning Versatile Convolution Filters for Efficient Visual Recognition
• Structured Pruning for Efficient ConvNets via Incremental Regularization
• Audio Interval Retrieval using Convolutional Neural Networks
• Broadcasted Residual Learning for Efficient Keyword Spotting
• Accelerating CNN Training by Pruning Activation Gradients
• Real-time Mask Detection on Google Edge TPU
• An Application-Specific VLIW Processor with Vector Instruction Set for CNN Acceleration
• FDFtNet: Facing Off Fake Images using Fake Detection Fine-tuning Network
• Real-time Person Re-identification at the Edge: A Mixed Precision Approach
• Tango: A Deep Neural Network Benchmark Suite for Various Accelerators
• A strong baseline for image and video quality assessment
• IRLAS: Inverse Reinforcement Learning for Architecture Search
• Making EfficientNet More Efficient: Exploring Batch-Independent Normalization, Group Convolutions and Reduced Resolution Training
• MiLeNAS: Efficient Neural Architecture Search via Mixed-Level Reformulation
• Efficient Anytime CLF Reactive Planning System for a Bipedal Robot on Undulating Terrain
• SparseRT: Accelerating Unstructured Sparsity on GPUs for Deep Learning Inference
• Learning Metrics from Teachers: Compact Networks for Image Embedding
• AttendNets: Tiny Deep Image Recognition Neural Networks for the Edge via Visual Attention Condensers
• PENNI: Pruned Kernel Sharing for Efficient CNN Inference
• flexgrid2vec: Learning Efficient Visual Representations Vectors
• A(DP)${}^2$SGD: Asynchronous Decentralized Parallel Stochastic Gradient Descent with Differential Privacy
• Characterising Across-Stack Optimisations for Deep Convolutional Neural Networks
• Collaborative Distillation for Ultra-Resolution Universal Style Transfer
• A Power-Efficient Binary-Weight Spiking Neural Network Architecture for Real-Time Object Classification
• Constrained deep neural network architecture search for IoT devices accounting hardware calibration
• Generalizable Pedestrian Detection: The Elephant In The Room
• Exploring Gradient Flow Based Saliency for DNN Model Compression
• Cascaded channel pruning using hierarchical self-distillation
• A Realtime Autonomous Robot Navigation Framework for Human like High-level Interaction and Task Planning in Global Dynamic Environment
• Orthogonal Convolutional Neural Networks
• Towards Lossless Binary Convolutional Neural Networks Using Piecewise Approximation
• VATLD: A Visual Analytics System to Assess, Understand and Improve Traffic Light Detection
• Progressive Neural Networks for Image Classification
• An Evaluation of Deep CNN Baselines for Scene-Independent Person Re-Identification
• Rethinking Floating Point Overheads for Mixed Precision DNN Accelerators
• MSplit LBI: Realizing Feature Selection and Dense Estimation Simultaneously in Few-shot and Zero-shot Learning
• An Features Extraction and Recognition Method for Underwater Acoustic Target Based on ATCNN
• Better the Devil you Know: An Analysis of Evasion Attacks using Out-of-Distribution Adversarial Examples
• FeatherNets: Convolutional Neural Networks as Light as Feather for Face Anti-spoofing
• Learning Whole-Image Descriptors for Real-time Loop Detection andKidnap Recovery under Large Viewpoint Difference
• Is Robustness the Cost of Accuracy? -- A Comprehensive Study on the Robustness of 18 Deep Image Classification Models
• A Facial Affect Analysis System for Autism Spectrum Disorder
• Deep Learning Towards Mobile Applications
• Real Time System for Facial Analysis
• Accelerating Minibatch Stochastic Gradient Descent using Typicality Sampling
• Learning from Higher-Layer Feature Visualizations
• Frustrated with Replicating Claims of a Shared Model? A Solution
• PointIT: A Fast Tracking Framework Based on 3D Instance Segmentation
• Fast Object Detection in Compressed Video
• Real-world Mapping of Gaze Fixations Using Instance Segmentation for Road Construction Safety Applications
• HSD-CNN: Hierarchically self decomposing CNN architecture using class specific filter sensitivity analysis
• A Near-Optimal Algorithm for Debiasing Trained Machine Learning Models
• MEAL: Manifold Embedding-based Active Learning
• Bring Your Own Codegen to Deep Learning Compiler
• MNN: A Universal and Efficient Inference Engine
• LE-HGR: A Lightweight and Efficient RGB-based Online Gesture Recognition Network for Embedded AR Devices
• Fully-Convolutional Intensive Feature Flow Neural Network for Text Recognition
• Manipulating SGD with Data Ordering Attacks
• Teachers Do More Than Teach: Compressing Image-to-Image Models
• A Novel Automation-Assisted Cervical Cancer Reading Method Based on Convolutional Neural Network
• Perceptual Extreme Super Resolution Network with Receptive Field Block
• ApproxDet: Content and Contention-Aware Approximate Object Detection for Mobiles
• Domain-Aware Dynamic Networks
• Reconfigurable Cyber-Physical System for Lifestyle Video-Monitoring via Deep Learning
• Faces à la Carte: Text-to-Face Generation via Attribute Disentanglement
• LRNNet: A Light-Weighted Network with Efficient Reduced Non-Local Operation for Real-Time Semantic Segmentation
• Grafted network for person re-identification
• Deep Learning for Efficient Reconstruction of High-Resolution Turbulent DNS Data
• InferBench: Understanding Deep Learning Inference Serving with an Automatic Benchmarking System
• Recurrent Neural Networks for video object detection
• Visual Localization for Autonomous Driving: Mapping the Accurate Location in the City Maze
• RobFR: Benchmarking Adversarial Robustness on Face Recognition
• DCANet: Learning Connected Attentions for Convolutional Neural Networks
• DrNAS: Dirichlet Neural Architecture Search
• Multi-Metric Evaluation of Thermal-to-Visual Face Recognition
• Learning Forward Reuse Distance
• Fitting the Search Space of Weight-sharing NAS with Graph Convolutional Networks
• A Lightweight Neural Network for Monocular View Generation with Occlusion Handling
• Real-time Memory Efficient Large-pose Face Alignment via Deep Evolutionary Network
• Observer Dependent Lossy Image Compression
• Efficient Semantic Scene Completion Network with Spatial Group Convolution
• HourNAS: Extremely Fast Neural Architecture Search Through an Hourglass Lens
• Real-time Detection, Tracking, and Classification of Moving and Stationary Objects using Multiple Fisheye Images
• Road User Detection in Videos
• Post-training Quantization with Multiple Points: Mixed Precision without Mixed Precision
• Improving Noise Robustness of an End-to-End Neural Model for Automatic Speech Recognition
• Global Context for Convolutional Pose Machines
• Auxiliary Learning for Deep Multi-task Learning
• Squeezed Convolutional Variational AutoEncoder for Unsupervised Anomaly Detection in Edge Device Industrial Internet of Things
• Towards Real-Time Monocular Depth Estimation for Robotics: A Survey
• Learning Student Networks via Feature Embedding
• COVID-19 personal protective equipment detection using real-time deep learning methods
• Evolving Search Space for Neural Architecture Search
• ESAI: Efficient Split Artificial Intelligence via Early Exiting Using Neural Architecture Search
• Neural Epitome Search for Architecture-Agnostic Network Compression
• How deep should be the depth of convolutional neural networks: a backyard dog case study
• Multi-task Graph Convolutional Neural Network for Calcification Morphology and Distribution Analysis in Mammograms
• Ordering Chaos: Memory-Aware Scheduling of Irregularly Wired Neural Networks for Edge Devices
• Real-Time Index Authentication for Event-Oriented Surveillance Video Query using Blockchain
• Faraway-Frustum: Dealing with Lidar Sparsity for 3D Object Detection using Fusion
• AutoKWS: Keyword Spotting with Differentiable Architecture Search
• Horizontally Fused Training Array: An Effective Hardware Utilization Squeezer for Training Novel Deep Learning Models
• Robust Real-time Pedestrian Detection in Aerial Imagery on Jetson TX2
• KS(conf ): A Light-Weight Test if a ConvNet Operates Outside of Its Specifications
• On the Privacy Risks of Cell-Based NAS Architectures
• Ternary Residual Networks
• Light-weight Document Image Cleanup using Perceptual Loss
• Nonuniform-to-Uniform Quantization: Towards Accurate Quantization via Generalized Straight-Through Estimation
• VarGFaceNet: An Efficient Variable Group Convolutional Neural Network for Lightweight Face Recognition
• PRGFlow: Benchmarking SWAP-Aware Unified Deep Visual Inertial Odometry
• Improved Detection of Adversarial Images Using Deep Neural Networks
• An Image Enhancing Pattern-based Sparsity for Real-time Inference on Mobile Devices
• Towards High Performance Video Object Detection
• AOGNets: Compositional Grammatical Architectures for Deep Learning
• Multiwavelength classification of X-ray selected galaxy cluster candidates using convolutional neural networks
• Camera View Adjustment Prediction for Improving Image Composition
• Towards large-scale, automated, accurate detection of CCTV camera objects using computer vision. Applications and implications for privacy, safety, and cybersecurity. (Preprint)
• Echo: Compiler-based GPU Memory Footprint Reduction for LSTM RNN Training
• Hire-MLP: Vision MLP via Hierarchical Rearrangement
• Embedded Large-Scale Handwritten Chinese Character Recognition
• Towards Deep and Efficient: A Deep Siamese Self-Attention Fully Efficient Convolutional Network for Change Detection in VHR Images
• Differentiable Dynamic Quantization with Mixed Precision and Adaptive Resolution
• Toward Efficient Transfer Learning in 6G
• A Semi-Automated Computational Approach for Infrared Dark Cloud Localization: A Catalog of Infrared Dark Clouds
• Application-driven Privacy-preserving Data Publishing with Correlated Attributes
• Condensation-Net: Memory-Efficient Network Architecture with Cross-Channel Pooling Layers and Virtual Feature Maps
• Validation of a deep learning mammography model in a population with low screening rates
• COP: Customized Deep Model Compression via Regularized Correlation-Based Filter-Level Pruning
• Refactoring Neural Networks for Verification
• ModuleNet: Knowledge-inherited Neural Architecture Search
• Energy-Efficient Adaptive Machine Learning on IoT End-Nodes With Class-Dependent Confidence
• Computation-Efficient Knowledge Distillation via Uncertainty-Aware Mixup
• SPARK: Spatial-aware Online Incremental Attack Against Visual Tracking
• Deep Dose Plugin Towards Real-time Monte Carlo Dose Calculation Through a Deep Learning based Denoising Algorithm
• Do Normalization Layers in a Deep ConvNet Really Need to Be Distinct?
• HM-NAS: Efficient Neural Architecture Search via Hierarchical Masking
• Multi Layer Neural Networks as Replacement for Pooling Operations
• A Highly Configurable Hardware/Software Stack for DNN Inference Acceleration
• Learning Strict Identity Mappings in Deep Residual Networks
• RILOD: Near Real-Time Incremental Learning for Object Detection at the Edge
• Hazard Detection in Supermarkets using Deep Learning on the Edge
• AgileNet: Lightweight Dictionary-based Few-shot Learning
• Interactive Classification for Deep Learning Interpretation
• DRINet: A Dual-Representation Iterative Learning Network for Point Cloud Segmentation
• Efficient Decoupled Neural Architecture Search by Structure and Operation Sampling
• Weight Pruning via Adaptive Sparsity Loss
• Seesaw-Net: Convolution Neural Network With Uneven Group Convolution
• Combinatorial Designs for Deep Learning
• Geometry-Aware Gradient Algorithms for Neural Architecture Search
• SGNet: A Super-class Guided Network for Image Classification and Object Detection
• Deep Attention Fusion Feature for Speech Separation with End-to-End Post-filter Method
• Differentiable Sparsification for Deep Neural Networks
• Automatic Neural Network Compression by Sparsity-Quantization Joint Learning: A Constrained Optimization-based Approach
• Exploiting Channel Similarity for Accelerating Deep Convolutional Neural Networks
• Hyperspectral Classification Based on 3D Asymmetric Inception Network with Data Fusion Transfer Learning
• Towards Characterizing Adversarial Defects of Deep Learning Software from the Lens of Uncertainty
• Fire SSD: Wide Fire Modules based Single Shot Detector on Edge Device
• Robust Training of Social Media Image Classification Models for Rapid Disaster Response
• DARC: Differentiable ARchitecture Compression
• A Pre-defined Sparse Kernel Based Convolution for Deep CNNs
• FlexSA: Flexible Systolic Array Architecture for Efficient Pruned DNN Model Training
• A Lightweight Structure Aimed to Utilize Spatial Correlation for Sparse-View CT Reconstruction
• Spatial-Separated Curve Rendering Network for Efficient and High-Resolution Image Harmonization
• Achieving Real-Time LiDAR 3D Object Detection on a Mobile Device
• Smart at what cost? Characterising Mobile Deep Neural Networks in the wild
• Task-Adaptive Neural Network Search with Meta-Contrastive Learning
• A Mean Field Theory of Quantized Deep Networks: The Quantization-Depth Trade-Off
• LightSAL: Lightweight Sign Agnostic Learning for Implicit Surface Representation
• Improving Accuracy of Binary Neural Networks using Unbalanced Activation Distribution
• Enhancing Model Assessment in Vision-based Interactive Machine Teaching through Real-time Saliency Map Visualization
• Report on UG^2+ Challenge Track 1: Assessing Algorithms to Improve Video Object Detection and Classification from Unconstrained Mobility Platforms
• CFPNet: Channel-wise Feature Pyramid for Real-Time Semantic Segmentation
• Mixed-Precision Quantized Neural Network with Progressively Decreasing Bitwidth For Image Classification and Object Detection
• Learning towards Minimum Hyperspherical Energy
• 2D or not 2D? Adaptive 3D Convolution Selection for Efficient Video Recognition
• NPAS: A Compiler-aware Framework of Unified Network Pruning and Architecture Search for Beyond Real-Time Mobile Acceleration
• A Simple Method to Reduce Off-chip Memory Accesses on Convolutional Neural Networks
• Scene Understanding Networks for Autonomous Driving based on Around View Monitoring System
• Building Proactive Voice Assistants: When and How (not) to Interact
• VA-RED${}^2$: Video Adaptive Redundancy Reduction
• Binarizing MobileNet via Evolution-based Searching
• Learning Compact Neural Networks Using Ordinary Differential Equations as Activation Functions
• Deep Scattering Network with Max-pooling
• Adaptive Selection of Deep Learning Models on Embedded Systems
• Automated Model Compression by Jointly Applied Pruning and Quantization
• Testing Deep Learning Models for Image Analysis Using Object-Relevant Metamorphic Relations
• Hessian-Aware Pruning and Optimal Neural Implant
• LP-3DCNN: Unveiling Local Phase in 3D Convolutional Neural Networks
• DKM: Differentiable K-Means Clustering Layer for Neural Network Compression
• Learning low-precision neural networks without Straight-Through Estimator(STE)
• Guidelines and Benchmarks for Deployment of Deep Learning Models on Smartphones as Real-Time Apps
• A Low-Cost Neural ODE with Depthwise Separable Convolution for Edge Domain Adaptation on FPGAs
• CALPA-NET: Channel-pruning-assisted Deep Residual Network for Steganalysis of Digital Images
• Integrating Large Circular Kernels into CNNs through Neural Architecture Search
• Towards Practical Lipreading with Distilled and Efficient Models
• MixSearch: Searching for Domain Generalized Medical Image Segmentation Architectures
• RGB-D Based Action Recognition with Light-weight 3D Convolutional Networks
• Low-Power Computer Vision: Status, Challenges, Opportunities
• Do All MobileNets Quantize Poorly? Gaining Insights into the Effect of Quantization on Depthwise Separable Convolutional Networks Through the Eyes of Multi-scale Distributional Dynamics
• Which *BERT? A Survey Organizing Contextualized Encoders
• FastFlowNet: A Lightweight Network for Fast Optical Flow Estimation
• Depthwise Spatio-Temporal STFT Convolutional Neural Networks for Human Action Recognition
• LPRNet: Lightweight Deep Network by Low-rank Pointwise Residual Convolution
• Generalized Data Weighting via Class-level Gradient Manipulation
• Implicit 3D Orientation Learning for 6D Object Detection from RGB Images
• Learning Architectures for Binary Networks
• GestARLite: An On-Device Pointing Finger Based Gestural Interface for Smartphones and Video See-Through Head-Mounts
• FreqNet: A Frequency-domain Image Super-Resolution Network with Dicrete Cosine Transform
• Contrastive Neural Architecture Search with Neural Architecture Comparators
• Dilated Convolution with Dilated GRU for Music Source Separation
• A Convolutional Neural Network-Based Low Complexity Filter
• Enabling Socially Competent navigation through incorporating HRI
• Hybrid Composition with IdleBlock: More Efficient Networks for Image Recognition
• DeepPeep: Exploiting Design Ramifications to Decipher the Architecture of Compact DNNs
• Adversarial Examples on Segmentation Models Can be Easy to Transfer
• AppealNet: An Efficient and Highly-Accurate Edge/Cloud Collaborative Architecture for DNN Inference
• BusyHands: A Hand-Tool Interaction Database for Assembly Tasks Semantic Segmentation
• ZARTS: On Zero-order Optimization for Neural Architecture Search
• MUXConv: Information Multiplexing in Convolutional Neural Networks
• MotionSqueeze: Neural Motion Feature Learning for Video Understanding
• DOTS: Decoupling Operation and Topology in Differentiable Architecture Search
• DPRed: Making Typical Activation and Weight Values Matter In Deep Learning Computing
• Building Efficient Deep Neural Networks with Unitary Group Convolutions
• Efficient Memory Management for Deep Neural Net Inference
• Pufferfish: Communication-efficient Models At No Extra Cost
• On the Distributional Properties of Adaptive Gradients
• BS-NAS: Broadening-and-Shrinking One-Shot NAS with Searchable Numbers of Channels
• AccelAT: A Framework for Accelerating the Adversarial Training of Deep Neural Networks through Accuracy Gradient
• Approximations in Deep Learning
• ML-EXray: Visibility into ML Deployment on the Edge
• Natural & Adversarial Bokeh Rendering via Circle-of-Confusion Predictive Network
• Energy-efficient Amortized Inference with Cascaded Deep Classifiers
• Leveraging Human Selective Attention for Medical Image Analysis with Limited Training Data
• Synthesis and Pruning as a Dynamic Compression Strategy for Efficient Deep Neural Networks
• PydMobileNet: Improved Version of MobileNets with Pyramid Depthwise Separable Convolution
• IntraQ: Learning Synthetic Images with Intra-Class Heterogeneity for Zero-Shot Network Quantization
• Inference of Recyclable Objects with Convolutional Neural Networks
• Learning to discover and localize visual objects with open vocabulary
• Fingerprint Spoof Buster
• A Survey of Mobile Computing for the Visually Impaired
• ESNet: An Efficient Symmetric Network for Real-time Semantic Segmentation
• Fully Learnable Group Convolution for Acceleration of Deep Neural Networks
• DeepLight: Learning Illumination for Unconstrained Mobile Mixed Reality
• A Unified Optimization Approach for CNN Model Inference on Integrated GPUs
• Block Convolution: Towards Memory-Efficient Inference of Large-Scale CNNs on FPGA
• AVA-Speech: A Densely Labeled Dataset of Speech Activity in Movies
• Dataflow-based Joint Quantization of Weights and Activations for Deep Neural Networks
• Layer Folding: Neural Network Depth Reduction using Activation Linearization
• Towards Real-Time Automatic Portrait Matting on Mobile Devices
• 3D Depthwise Convolution: Reducing Model Parameters in 3D Vision Tasks
• Locally Enhanced Self-Attention: Combining Self-Attention and Convolution as Local and Context Terms
• M-FAC: Efficient Matrix-Free Approximations of Second-Order Information
• Matrix and tensor decompositions for training binary neural networks
• KVT: k-NN Attention for Boosting Vision Transformers
• Mixup Regularization for Region Proposal based Object Detectors
• Searching for Fast Model Families on Datacenter Accelerators
• Dynamic-OFA: Runtime DNN Architecture Switching for Performance Scaling on Heterogeneous Embedded Platforms
• SLSGD: Secure and Efficient Distributed On-device Machine Learning
• Neural Architecture Search for Lightweight Non-Local Networks
• Effective Sparsification of Neural Networks with Global Sparsity Constraint
• Depth Quality-Inspired Feature Manipulation for Efficient RGB-D Salient Object Detection
• CrossStack: A 3-D Reconfigurable RRAM Crossbar Inference Engine
• E-PixelHop: An Enhanced PixelHop Method for Object Classification
• Neural Machine Translation with Joint Representation
• A New Clustering-Based Technique for the Acceleration of Deep Convolutional Networks
• GroupReduce: Block-Wise Low-Rank Approximation for Neural Language Model Shrinking
• On Machine Learning and Structure for Mobile Robots
• Collage Inference: Using Coded Redundancy for Low Variance Distributed Image Classification
• Resource Constrained Neural Network Architecture Search: Will a Submodularity Assumption Help?
• Block Annotation: Better Image Annotation for Semantic Segmentation with Sub-Image Decomposition
• Semi-Supervised Segmentation of Concrete Aggregate Using Consensus Regularisation and Prior Guidance
• Anomaly Detection in Residential Video Surveillance on Edge Devices in IoT Framework
• EdgeSegNet: A Compact Network for Semantic Segmentation
• Predictive Visual Tracking: A New Benchmark and Baseline Approach
• DeepWear: Adaptive Local Offloading for On-Wearable Deep Learning
• Slimmable Compressive Autoencoders for Practical Neural Image Compression
• ForgeryNet: A Versatile Benchmark for Comprehensive Forgery Analysis
• Predicting Terrain Mechanical Properties in Sight for Planetary Rovers with Semantic Clues
• AlphaNet: Improved Training of Supernets with Alpha-Divergence
• SAIA: Split Artificial Intelligence Architecture for Mobile Healthcare System
• The Impact of Reinitialization on Generalization in Convolutional Neural Networks
• Low-Cost Parameterizations of Deep Convolutional Neural Networks
• Parallel Residual Bi-Fusion Feature Pyramid Network for Accurate Single-Shot Object Detection
• Training Meta-Surrogate Model for Transferable Adversarial Attack
• Neural Pruning via Growing Regularization
• Jointly Optimizing Preprocessing and Inference for DNN-based Visual Analytics
• Learning to Fuse Asymmetric Feature Maps in Siamese Trackers
• SplitSR: An End-to-End Approach to Super-Resolution on Mobile Devices
• Single-Path Mobile AutoML: Efficient ConvNet Design and NAS Hyperparameter Optimization
• On Benchmarking Iris Recognition within a Head-mounted Display for AR/VR Application
• SeesawFaceNets: sparse and robust face verification model for mobile platform
• Learning Efficient Video Representation with Video Shuffle Networks
• MoGA: Searching Beyond MobileNetV3
• FTPipeHD: A Fault-Tolerant Pipeline-Parallel Distributed Training Framework for Heterogeneous Edge Devices
• Fingerprinting Multi-exit Deep Neural Network Models via Inference Time
• DC-NAS: Divide-and-Conquer Neural Architecture Search
• End-to-end Interpretable Neural Motion Planner
• Evaluating Off-the-Shelf Machine Listening and Natural Language Models for Automated Audio Captioning
• Open DNN Box by Power Side-Channel Attack
• Randomized Overdrive Neural Networks
• Enhancing a Neurocognitive Shared Visuomotor Model for Object Identification, Localization, and Grasping With Learning From Auxiliary Tasks
• Automated Rib Fracture Detection of Postmortem Computed Tomography Images Using Machine Learning Techniques
• Comprehensive Comparison of Deep Learning Models for Lung and COVID-19 Lesion Segmentation in CT scans
• DeepOrganNet: On-the-Fly Reconstruction and Visualization of 3D / 4D Lung Models from Single-View Projections by Deep Deformation Network
• Dense Residual Network: Enhancing Global Dense Feature Flow for Character Recognition
• Binarized Neural Architecture Search for Efficient Object Recognition
• TanhSoft -- a family of activation functions combining Tanh and Softplus
• Fine-Grained Stochastic Architecture Search
• Frame-To-Frame Consistent Semantic Segmentation
• Discovering Multi-Hardware Mobile Models via Architecture Search
• Self-Supervised Learning of a Biologically-Inspired Visual Texture Model
• FlatteNet: A Simple Versatile Framework for Dense Pixelwise Prediction
• Knowledge Transfer via Dense Cross-Layer Mutual-Distillation
• FDDWNet: A Lightweight Convolutional Neural Network for Real-time Sementic Segmentation
• EfficientHRNet: Efficient Scaling for Lightweight High-Resolution Multi-Person Pose Estimation
• LCP: A Low-Communication Parallelization Method for Fast Neural Network Inference in Image Recognition
• MutualNet: Adaptive ConvNet via Mutual Learning from Network Width and Resolution
• Training Certifiably Robust Neural Networks with Efficient Local Lipschitz Bounds
• Steepest Descent Neural Architecture Optimization: Escaping Local Optimum with Signed Neural Splitting
• Dynamic Group Convolution for Accelerating Convolutional Neural Networks
• VACL: Variance-Aware Cross-Layer Regularization for Pruning Deep Residual Networks
• Orthant Based Proximal Stochastic Gradient Method for ${\ell }_1$-Regularized Optimization
• FNA++: Fast Network Adaptation via Parameter Remapping and Architecture Search
• Dual-attention Focused Module for Weakly Supervised Object Localization
• MobileDepth: Efficient Monocular Depth Prediction on Mobile Devices
• A Matrix-in-matrix Neural Network for Image Super Resolution
• Distilling with Performance Enhanced Students
• REPrune: Filter Pruning via Representative Election
• Skin disease diagnosis with deep learning: a review
• Convolutional Neural Network Quantization using Generalized Gamma Distribution
• Hidden-Fold Networks: Random Recurrent Residuals Using Sparse Supermasks
• PAM: Pose Attention Module for Pose-Invariant Face Recognition
• STH: Spatio-Temporal Hybrid Convolution for Efficient Action Recognition
• Understanding and Testing Generalization of Deep Networks on Out-of-Distribution Data
• Improving Binary Neural Networks through Fully Utilizing Latent Weights
• SECS: Efficient Deep Stream Processing via Class Skew Dichotomy
• FrostNet: Towards Quantization-Aware Network Architecture Search
• Proximu$: Efficiently Scaling DNN Inference in Multi-core CPUs through Near-Cache Compute
• Gated Multi-layer Convolutional Feature Extraction Network for Robust Pedestrian Detection
• Disturbance-immune Weight Sharing for Neural Architecture Search
• WSNet: Compact and Efficient Networks Through Weight Sampling
• Automated Quality Assessment of Hand Washing Using Deep Learning
• MicroNet: Improving Image Recognition with Extremely Low FLOPs
• Joint Estimation of Age and Gender from Unconstrained Face Images using Lightweight Multi-task CNN for Mobile Applications
• Brain-inspired reverse adversarial examples
• Developing a Compressed Object Detection Model based on YOLOv4 for Deployment on Embedded GPU Platform of Autonomous System
• Refining the Structure of Neural Networks Using Matrix Conditioning
• A Real-time Low-cost Artificial Intelligence System for Autonomous Spraying in Palm Plantations
• Malware Classification Using Transfer Learning
• Efficient CNN Building Blocks for Encrypted Data
• AdaDeep: A Usage-Driven, Automated Deep Model Compression Framework for Enabling Ubiquitous Intelligent Mobiles
• AttendSeg: A Tiny Attention Condenser Neural Network for Semantic Segmentation on the Edge
• Parameter Prediction for Unseen Deep Architectures
• Sisyphus: A Cautionary Tale of Using Low-Degree Polynomial Activations in Privacy-Preserving Deep Learning
• Optimizing Neural Architecture Search using Limited GPU Time in a Dynamic Search Space: A Gene Expression Programming Approach
• Fast-Tracker 2.0: Improving Autonomy of Aerial Tracking with Active Vision and Human Location Regression
• HALP: Hardware-Aware Latency Pruning
• An Evolution of CNN Object Classifiers on Low-Resolution Images
• HASCO: Towards Agile HArdware and Software CO-design for Tensor Computation
• Model-inspired Deep Learning for Light-Field Microscopy with Application to Neuron Localization
• Make (Nearly) Every Neural Network Better: Generating Neural Network Ensembles by Weight Parameter Resampling
• TensorFlow with user friendly Graphical Framework for object detection API
• TENET: A Framework for Modeling Tensor Dataflow Based on Relation-centric Notation
• Full-attention based Neural Architecture Search using Context Auto-regression
• Differentiable Learning-to-Group Channels via Groupable Convolutional Neural Networks
• Adversarial Attack across Datasets
• Automated flow for compressing convolution neural networks for efficient edge-computation with FPGA
• NetAdaptV2: Efficient Neural Architecture Search with Fast Super-Network Training and Architecture Optimization
• AdaptCL: Efficient Collaborative Learning with Dynamic and Adaptive Pruning
• Salienteye: Maximizing Engagement While Maintaining Artistic Style on Instagram Using Deep Neural Networks
• ACDC: Weight Sharing in Atom-Coefficient Decomposed Convolution
• Optimal Quantization for Batch Normalization in Neural Network Deployments and Beyond
• Tracking e-cigarette warning label compliance on Instagram with deep learning
• A free web service for fast COVID-19 classification of chest X-Ray images
• Decentralized Smart Surveillance through Microservices Platform
• Dataflow Aware Mapping of Convolutional Neural Networks Onto Many-Core Platforms With Network-on-Chip Interconnect
• Table-Based Neural Units: Fully Quantizing Networks for Multiply-Free Inference
• Efficient Integer-Arithmetic-Only Convolutional Neural Networks
• n-hot: Efficient bit-level sparsity for powers-of-two neural network quantization
• FurcaNeXt: End-to-end monaural speech separation with dynamic gated dilated temporal convolutional networks
• Learning compact generalizable neural representations supporting perceptual grouping
• On the Learning Property of Logistic and Softmax Losses for Deep Neural Networks
• FRDet: Balanced and Lightweight Object Detector based on Fire-Residual Modules for Embedded Processor of Autonomous Driving
• Knowledge Distillation via Instance-level Sequence Learning
• AnalogNet: Convolutional Neural Network Inference on Analog Focal Plane Sensor Processors
• Deep Neural Models for color discrimination and color constancy
• When to Prune? A Policy towards Early Structural Pruning
• Improved Speech Separation with Time-and-Frequency Cross-domain Joint Embedding and Clustering
• Training convolutional neural networks with cheap convolutions and online distillation
• Split to Be Slim: An Overlooked Redundancy in Vanilla Convolution
• Learning Sparse Mixture of Experts for Visual Question Answering
• Learning Sparse & Ternary Neural Networks with Entropy-Constrained Trained Ternarization (EC2T)
• Pyramidal Dense Attention Networks for Lightweight Image Super-Resolution
• Dynamic Filtering with Large Sampling Field for ConvNets
• Joint Architecture and Knowledge Distillation in CNN for Chinese Text Recognition
• Principal Component Networks: Parameter Reduction Early in Training
• GAN-Knowledge Distillation for one-stage Object Detection
• Multi-task Learning with Attention for End-to-end Autonomous Driving
• Scheduled Differentiable Architecture Search for Visual Recognition
• Stacked Temporal Attention: Improving First-person Action Recognition by Emphasizing Discriminative Clips
• Separable Layers Enable Structured Efficient Linear Substitutions
• Improving On-Screen Sound Separation for Open-Domain Videos with Audio-Visual Self-Attention
• ASAP-NMS: Accelerating Non-Maximum Suppression Using Spatially Aware Priors
• Capsule network with shortcut routing
• TYolov5: A Temporal Yolov5 Detector Based on Quasi-Recurrent Neural Networks for Real-Time Handgun Detection in Video
• Swift for TensorFlow: A portable, flexible platform for deep learning
• Beyond Self-Supervision: A Simple Yet Effective Network Distillation Alternative to Improve Backbones
• Feature Products Yield Efficient Networks
• Improving Sickle Cell Disease Classification: A Fusion of Conventional Classifiers, Segmented Images, and Convolutional Neural Networks
• MVStylizer: An Efficient Edge-Assisted Video Photorealistic Style Transfer System for Mobile Phones
• Increasing Trustworthiness of Deep Neural Networks via Accuracy Monitoring
• Graph-Adaptive Pruning for Efficient Inference of Convolutional Neural Networks
• Train-by-Reconnect: Decoupling Locations of Weights from their Values
• Towards Efficient Convolutional Neural Network for Domain-Specific Applications on FPGA
• Binarized Neural Architecture Search
• HENet:A Highly Efficient Convolutional Neural Networks Optimized for Accuracy, Speed and Storage
• On Learning Semantic Representations for Million-Scale Free-Hand Sketches
• SlideNet: Fast and Accurate Slide Quality Assessment Based on Deep Neural Networks
• PatchFormer: An Efficient Point Transformer with Patch Attention
• Binary Input Layer: Training of CNN models with binary input data
• A First Look at Deep Learning Apps on Smartphones
• ACP: Automatic Channel Pruning via Clustering and Swarm Intelligence Optimization for CNN
• SparseDNN: Fast Sparse Deep Learning Inference on CPUs
• ResNetX: a more disordered and deeper network architecture
• ECG-DelNet: Delineation of Ambulatory Electrocardiograms with Mixed Quality Labeling Using Neural Networks
• Analyzing Machine Learning Workloads Using a Detailed GPU Simulator
• deepSELF: An Open Source Deep Self End-to-End Learning Framework
• Artificial Intelligence for COVID-19 Detection -- A state-of-the-art review
• CodeVIO: Visual-Inertial Odometry with Learned Optimizable Dense Depth
• ViPNAS: Efficient Video Pose Estimation via Neural Architecture Search
• LeanResNet: A Low-cost Yet Effective Convolutional Residual Networks
• FALCON: Lightweight and Accurate Convolution
• A machine learning environment for evaluating autonomous driving software
• Distributed Low Precision Training Without Mixed Precision
• Edge-Cloud Collaborated Object Detection via Difficult-Case Discriminator
• S2-BNN: Bridging the Gap Between Self-Supervised Real and 1-bit Neural Networks via Guided Distribution Calibration
• How Unique Is a Face: An Investigative Study
• MemNet: Memory-Efficiency Guided Neural Architecture Search with Augment-Trim learning
• ConvNets for Counting: Object Detection of Transient Phenomena in Steelpan Drums
• Correlation Congruence for Knowledge Distillation
• TE-YOLOF: Tiny and efficient YOLOF for blood cell detection
• Efficient Integration of Multi-channel Information for Speaker-independent Speech Separation
• CAM-GAN: Continual Adaptation Modules for Generative Adversarial Networks
• Fine-grained Data Distribution Alignment for Post-Training Quantization
• HGC: Hierarchical Group Convolution for Highly Efficient Neural Network
• ECC: Platform-Independent Energy-Constrained Deep Neural Network Compression via a Bilinear Regression Model
• 3D Dense Separated Convolution Module for Volumetric Image Analysis
• Face Recognition in Unconstrained Conditions: A Systematic Review
• Data-Driven Compression of Convolutional Neural Networks
• Embedded Knowledge Distillation in Depth-Level Dynamic Neural Network
• InfantNet: A Deep Neural Network for Analyzing Infant Vocalizations
• Dynamic Compression Ratio Selection for Edge Inference Systems with Hard Deadlines
• GANMEX: One-vs-One Attributions Guided by GAN-based Counterfactual Explanation Baselines
• Live Target Detection with Deep Learning Neural Network and Unmanned Aerial Vehicle on Android Mobile Device
• Propose-and-Attend Single Shot Detector
• WeightNet: Revisiting the Design Space of Weight Networks
• Parameter Efficient Deep Neural Networks with Bilinear Projections
• DS-Net++: Dynamic Weight Slicing for Efficient Inference in CNNs and Transformers
• Efficient Pig Counting in Crowds with Keypoints Tracking and Spatial-aware Temporal Response Filtering
• On-Device Document Classification using multimodal features
• EPNAS: Efficient Progressive Neural Architecture Search
• Deep multi-modal networks for book genre classification based on its cover
• High Performance Depthwise and Pointwise Convolutions on Mobile Devices
• P3SGD: Patient Privacy Preserving SGD for Regularizing Deep CNNs in Pathological Image Classification
• Distribution-sensitive Information Retention for Accurate Binary Neural Network
• A Demonstration of Smart Doorbell Design Using Federated Deep Learning
• Learning Depthwise Separable Graph Convolution from Data Manifold
• PointAR: Efficient Lighting Estimation for Mobile Augmented Reality
• ExplainFix: Explainable Spatially Fixed Deep Networks
• DeepDive: An Integrative Algorithm/Architecture Co-Design for Deep Separable Convolutional Neural Networks
• Merging and Evolution: Improving Convolutional Neural Networks for Mobile Applications
• Resource Rationing for Wireless Federated Learning: Concept, Benefits, and Challenges
• Recent Advances in Efficient Computation of Deep Convolutional Neural Networks
• Progressive Learning of Low-Precision Networks
• Efficient CNN-LSTM based Image Captioning using Neural Network Compression
• Developing efficient transfer learning strategies for robust scene recognition in mobile robotics using pre-trained convolutional neural networks
• Where Should We Begin? A Low-Level Exploration of Weight Initialization Impact on Quantized Behaviour of Deep Neural Networks
• The Unreasonable Effectiveness of Encoder-Decoder Networks for Retinal Vessel Segmentation
• Investigating Emotion-Color Association in Deep Neural Networks
• SuperOCR: A Conversion from Optical Character Recognition to Image Captioning
• Real-Time Edge Classification: Optimal Offloading under Token Bucket Constraints
• Joint Channel and Weight Pruning for Model Acceleration on Moblie Devices
• DPNET: Dual-Path Network for Efficient Object Detectioj with Lightweight Self-Attention
• Learning OFDM Waveforms with PAPR and ACLR Constraints
• BNAS v2: Learning Architectures for Binary Networks with Empirical Improvements
• Advances and Challenges in Deep Lip Reading
• LCS: Learning Compressible Subspaces for Adaptive Network Compression at Inference Time
• Pre-training without Natural Images
• Exploiting Activation based Gradient Output Sparsity to Accelerate Backpropagation in CNNs
• Convolutional Hough Matching Networks for Robust and Efficient Visual Correspondence
• Performance Evaluation of Convolutional Neural Networks for Gait Recognition
• Deep Neural Networks for Active Wave Breaking Classification
• An Attention Module for Convolutional Neural Networks
• AIRCHITECT: Learning Custom Architecture Design and Mapping Space
• VTLayout: Fusion of Visual and Text Features for Document Layout Analysis
• Real-time Keypoints Detection for Autonomous Recovery of the Unmanned Ground Vehicle
• BinArray: A Scalable Hardware Accelerator for Binary Approximated CNNs
• Follow Your Path: a Progressive Method for Knowledge Distillation
• Robustness of on-device Models: Adversarial Attack to Deep Learning Models on Android Apps
• Universal Adder Neural Networks
• Practical Assessment of Generalization Performance Robustness for Deep Networks via Contrastive Examples
• GroupBERT: Enhanced Transformer Architecture with Efficient Grouped Structures
• Face mask detection using convolution neural network
• MLPerf Tiny Benchmark
• Differentiable Neural Architecture Learning for Efficient Neural Network Design
• Concurrent Neural Tree and Data Preprocessing AutoML for Image Classification
• S4Net: Single Stage Salient-Instance Segmentation
• Sub-pixel face landmarks using heatmaps and a bag of tricks
• A Fast Knowledge Distillation Framework for Visual Recognition
• PatchNet -- Short-range Template Matching for Efficient Video Processing
• Interleaving Learning, with Application to Neural Architecture Search
• Decentralized Reinforcement Learning for Multi-Target Search and Detection by a Team of Drones
• Detecting Interlocutor Confusion in Situated Human-Avatar Dialogue: A Pilot Study
• SwiftSRGAN -- Rethinking Super-Resolution for Efficient and Real-time Inference
• Video Frame Interpolation Transformer
• Fast DCTTS: Efficient Deep Convolutional Text-to-Speech
• The VVAD-LRS3 Dataset for Visual Voice Activity Detection
• Efficient Action Recognition Using Confidence Distillation
• Incomplete Dot Products for Dynamic Computation Scaling in Neural Network Inference
• Humans as Path-Finders for Safe Navigation
• Towards Efficient Convolutional Network Models with Filter Distribution Templates
• TimeGate: Conditional Gating of Segments in Long-range Activities
• MODS -- A USV-oriented object detection and obstacle segmentation benchmark
• Less is More: Accelerating Faster Neural Networks Straight from JPEG
• IROS 2019 Lifelong Robotic Vision Challenge -- Lifelong Object Recognition Report
• Device-Cloud Collaborative Learning for Recommendation
• iToF2dToF: A Robust and Flexible Representation for Data-Driven Time-of-Flight Imaging
• Low Dimensional Landscape Hypothesis is True: DNNs can be Trained in Tiny Subspaces
• ReCU: Reviving the Dead Weights in Binary Neural Networks
• Active multi-fidelity Bayesian online changepoint detection
• SiMaN: Sign-to-Magnitude Network Binarization
• Towards Streaming Perception
• Seismic Shot Gather Noise Localization Using a Multi-Scale Feature-Fusion-Based Neural Network
• MLPerf Mobile Inference Benchmark
• ShadowNet: A Secure and Efficient On-device Model Inference System for Convolutional Neural Networks
• Ranking Neural Checkpoints
• Deep Learning in the Era of Edge Computing: Challenges and Opportunities
• Post-Training BatchNorm Recalibration
• Adversarial Robustness through Bias Variance Decomposition: A New Perspective for Federated Learning
• Half-Space Proximal Stochastic Gradient Method for Group-Sparsity Regularized Problem
• Little Motion, Big Results: Using Motion Magnification to Reveal Subtle Tremors in Infants
• Training Sparse Neural Networks using Compressed Sensing
• SAFRON: Stitching Across the Frontier for Generating Colorectal Cancer Histology Images
• A Light-Weighted Convolutional Neural Network for Bitemporal SAR Image Change Detection
• Matching Guided Distillation
• AQD: Towards Accurate Fully-Quantized Object Detection
• On the Demystification of Knowledge Distillation: A Residual Network Perspective
• Representation Sharing for Fast Object Detector Search and Beyond
• Fast Object Classification and Meaningful Data Representation of Segmented Lidar Instances
• Client Adaptation improves Federated Learning with Simulated Non-IID Clients
• Knowledge Distillation Meets Self-Supervision
• SADet: Learning An Efficient and Accurate Pedestrian Detector
• A Survey on Edge Performance Benchmarking
• Non-Blocking Simultaneous Multithreading: Embracing the Resiliency of Deep Neural Networks
• MDInference: Balancing Inference Accuracy and Latency for Mobile Applications
• Evaluation of Model Selection for Kernel Fragment Recognition in Corn Silage
• DA-NAS: Data Adapted Pruning for Efficient Neural Architecture Search
• The 1st Challenge on Remote Physiological Signal Sensing (RePSS)
• Cross-Model Image Annotation Platform with Active Learning
• A Survey on Large-scale Machine Learning
• Performance Evaluation of Low-Cost Machine Vision Cameras for Image-Based Grasp Verification
• Representative Graph Neural Network
• Weight Equalizing Shift Scaler-Coupled Post-training Quantization
• Channel Pruning via Optimal Thresholding
• Exploiting Fully Convolutional Network and Visualization Techniques on Spontaneous Speech for Dementia Detection
• ResiliNet: Failure-Resilient Inference in Distributed Neural Networks
• Extending Label Smoothing Regularization with Self-Knowledge Distillation
• Margin-Based Regularization and Selective Sampling in Deep Neural Networks
• On the Orthogonality of Knowledge Distillation with Other Techniques: From an Ensemble Perspective
• ENAS4D: Efficient Multi-stage CNN Architecture Search for Dynamic Inference
• Coupled Network for Robust Pedestrian Detection with Gated Multi-Layer Feature Extraction and Deformable Occlusion Handling
• Accelerating Multi-Model Inference by Merging DNNs of Different Weights
• Neural Machine Translation: A Review and Survey
• TinyGAN: Distilling BigGAN for Conditional Image Generation
• Eye Semantic Segmentation with a Lightweight Model
• ISP4ML: Understanding the Role of Image Signal Processing in Efficient Deep Learning Vision Systems
• Balancing Specialization, Generalization, and Compression for Detection and Tracking
• SPEC2: SPECtral SParsE CNN Accelerator on FPGAs
• Searching for Accurate Binary Neural Architectures
• DASNet: Dynamic Activation Sparsity for Neural Network Efficiency Improvement
• Understanding the Effects of Pre-Training for Object Detectors via Eigenspectrum
• ModiPick: SLA-aware Accuracy Optimization For Mobile Deep Inference
• Thanks for Nothing: Predicting Zero-Valued Activations with Lightweight Convolutional Neural Networks
• Robust Membership Encoding: Inference Attacks and Copyright Protection for Deep Learning
• The Pitfall of Evaluating Performance on Emerging AI Accelerators
• Linear Context Transform Block
• Adversarial-Based Knowledge Distillation for Multi-Model Ensemble and Noisy Data Refinement
• HBONet: Harmonious Bottleneck on Two Orthogonal Dimensions
• Explaining Image Classifiers using Statistical Fault Localization
• Tuning Algorithms and Generators for Efficient Edge Inference
• Hand-Gesture-Recognition Based Text Input Method for AR/VR Wearable Devices
• MITAS: A Compressed Time-Domain Audio Separation Network with Parameter Sharing
• RTN: Reparameterized Ternary Network
• RecNets: Channel-wise Recurrent Convolutional Neural Networks
• Small, Accurate, and Fast Vehicle Re-ID on the Edge: the SAFR Approach
• Improving Efficiency in Neural Network Accelerator Using Operands Hamming Distance optimization
• Self-supervised learning for audio-visual speaker diarization
• Multi-Task Multicriteria Hyperparameter Optimization
• DotFAN: A Domain-transferred Face Augmentation Network for Pose and Illumination Invariant Face Recognition
• Dynamic Spatial Verification for Large-Scale Object-Level Image Retrieval
• Understanding Chat Messages for Sticker Recommendation in Messaging Apps
• URNet : User-Resizable Residual Networks with Conditional Gating Module
• Doubly Sparse: Sparse Mixture of Sparse Experts for Efficient Softmax Inference
• Exploring Weight Symmetry in Deep Neural Networks
• Fast Adjustable Threshold For Uniform Neural Network Quantization (Winning solution of LPIRC-II)
• WaveletNet: Logarithmic Scale Efficient Convolutional Neural Networks for Edge Devices
• Dense xUnit Networks
• Human Face Expressions from Images - 2D Face Geometry and 3D Face Local Motion versus Deep Neural Features
• A Simple Non-i.i.d. Sampling Approach for Efficient Training and Better Generalization
• TEA-DNN: the Quest for Time-Energy-Accuracy Co-optimized Deep Neural Networks
• Accelerating Training of Deep Neural Networks with a Standardization Loss
• A lightweight convolutional neural network for image denoising with fine details preservation capability
• Real-time Egocentric Gesture Recognition on Mobile Head Mounted Displays
• Vehicle Re-Identification in Context
• Towards a Generic Diver-Following Algorithm: Balancing Robustness and Efficiency in Deep Visual Detection
• Sparsity in Deep Neural Networks - An Empirical Investigation with TensorQuant
• Kerman: A Hybrid Lightweight Tracking Algorithm to Enable Smart Surveillance as an Edge Service
• Efficient Uncertainty Estimation for Semantic Segmentation in Videos
• Compression and Localization in Reinforcement Learning for ATARI Games
• Efficient Semantic Segmentation using Gradual Grouping
• High Frequency Residual Learning for Multi-Scale Image Classification
• Massively Parallel Video Networks
• MPDCompress - Matrix Permutation Decomposition Algorithm for Deep Neural Network Compression
• Ultra Power-Efficient CNN Domain Specific Accelerator with 9.3TOPS/Watt for Mobile and Embedded Applications
• Parameter Transfer Unit for Deep Neural Networks
• Data Analytics Service Composition and Deployment on Edge Devices
• Diagonalwise Refactorization: An Efficient Training Method for Depthwise Convolutions
• Hardware-friendly Neural Network Architecture for Neuromorphic Computing
• Temporally Identity-Aware SSD with Attentional LSTM
• Fine-Grained Energy and Performance Profiling framework for Deep Convolutional Neural Networks
• MobiVSR: A Visual Speech Recognition Solution for Mobile Devices
• Object Detection with Mask-based Feature Encoding
• Genetic Network Architecture Search
• Model Optimization for Deep Space Exploration via Simulators and Deep Learning
• Exploring Neural Networks Quantization via Layer-Wise Quantization Analysis
• Loss Landscape Dependent Self-Adjusting Learning Rates in Decentralized Stochastic Gradient Descent
• A VM/Containerized Approach for Scaling TinyML Applications
• Perception Framework through Real-Time Semantic Segmentation and Scene Recognition on a Wearable System for the Visually Impaired
• Protecting Geolocation Privacy of Photo Collections
• A multimodal deep learning framework for scalable content based visual media retrieval
• Analytical aspects of non-differentiable neural networks
• ShortcutFusion: From Tensorflow to FPGA-based accelerator with reuse-aware memory allocation for shortcut data
• Multi-level Feature Fusion-based CNN for Local Climate Zone Classification from Sentinel-2 Images: Benchmark Results on the So2Sat LCZ42 Dataset
• All-You-Can-Fit 8-Bit Flexible Floating-Point Format for Accurate and Memory-Efficient Inference of Deep Neural Networks
• Zero-Cost Operation Scoring in Differentiable Architecture Search
• Studying the Plasticity in Deep Convolutional Neural Networks using Random Pruning
• PAL: Intelligence Augmentation using Egocentric Visual Context Detection
• Multi-level Knowledge Distillation via Knowledge Alignment and Correlation
• cofga: A Dataset for Fine Grained Classification of Objects from Aerial Imagery
• Event Recognition with Automatic Album Detection based on Sequential Processing, Neural Attention and Image Captioning
• OD-SGD: One-step Delay Stochastic Gradient Descent for Distributed Training
• COFGA: Classification Of Fine-Grained Features In Aerial Images
• Estimating the Robustness of Classification Models by the Structure of the Learned Feature-Space
• Group Pruning using a Bounded-Lp norm for Group Gating and Regularization
• CamLoc: Pedestrian Location Detection from Pose Estimation on Resource-constrained Smart-cameras
• Finet: Using Fine-grained Batch Normalization to Train Light-weight Neural Networks
• Filter Bank Regularization of Convolutional Neural Networks
• Multi-Scale Temporal Convolution Network for Classroom Voice Detection
• Efficient Human Pose Estimation with Depthwise Separable Convolution and Person Centroid Guided Joint Grouping
• Joint Matrix Decomposition for Deep Convolutional Neural Networks Compression
• Greenery Segmentation In Urban Images By Deep Learning
• Dynamic Resolution Network
• Personalizing Pre-trained Models
• IEA: Inner Ensemble Average within a convolutional neural network
• Energy Drain of the Object Detection Processing Pipeline for Mobile Devices: Analysis and Implications
• Towards Accurate and Compact Architectures via Neural Architecture Transformer
• Road Segmentation Using CNN and Distributed LSTM
• Live Reconstruction of Large-Scale Dynamic Outdoor Worlds
• Robust Student Network Learning
• Self-Reorganizing and Rejuvenating CNNs for Increasing Model Capacity Utilization
• "BNN - BN = ?": Training Binary Neural Networks without Batch Normalization
• Global Context Networks
• Channel-wise pruning of neural networks with tapering resource constraint
• Neural Rejuvenation: Improving Deep Network Training by Enhancing Computational Resource Utilization
• Multigrid-in-Channels Architectures for Wide Convolutional Neural Networks
• RoadNet-RT: High Throughput CNN Architecture and SoC Design for Real-Time Road Segmentation
• Person Identification with Visual Summary for a Safe Access to a Smart Home
• Metric Embedding Autoencoders for Unsupervised Cross-Dataset Transfer Learning
• Learning to Cascade: Confidence Calibration for Improving the Accuracy and Computational Cost of Cascade Inference Systems
• BFTrainer: Low-Cost Training of Neural Networks on Unfillable Supercomputer Nodes
• Augmentation Inside the Network
• ESFNet: Efficient Network for Building Extraction from High-Resolution Aerial Images
• Multi-objective Neural Architecture Search with Almost No Training
• Greedy Network Enlarging
• Spending Your Winning Lottery Better After Drawing It
• Full-Stack Filters to Build Minimum Viable CNNs
• A Visual Domain Transfer Learning Approach for Heartbeat Sound Classification
• Rapid Elastic Architecture Search under Specialized Classes and Resource Constraints
• MS-DARTS: Mean-Shift Based Differentiable Architecture Search
• PSDNet and DPDNet: Efficient channel expansion, Depthwise-Pointwise-Depthwise Inverted Bottleneck Block
• Fully Quantized Image Super-Resolution Networks
• Neural Architecture Search as Sparse Supernet
• SkyScapes -- Fine-Grained Semantic Understanding of Aerial Scenes
• A System-Level Solution for Low-Power Object Detection
• NeuralScale: Efficient Scaling of Neurons for Resource-Constrained Deep Neural Networks
• Light-weighted Saliency Detection with Distinctively Lower Memory Cost and Model Size
• IMMVP: An Efficient Daytime and Nighttime On-Road Object Detector
• IC Networks: Remodeling the Basic Unit for Convolutional Neural Networks
• Deep Sensing of Urban Waterlogging
• Deployment of Customized Deep Learning based Video Analytics On Surveillance Cameras
• Training for temporal sparsity in deep neural networks, application in video processing
• CondenseNet V2: Sparse Feature Reactivation for Deep Networks
• FactorizeNet: Progressive Depth Factorization for Efficient Network Architecture Exploration Under Quantization Constraints
• An Artificial Intelligence System for Combined Fruit Detection and Georeferencing, Using RTK-Based Perspective Projection in Drone Imagery
• A Machine Learning Imaging Core using Separable FIR-IIR Filters
• Semantically Selective Augmentation for Deep Compact Person Re-Identification
• The Elastic Lottery Ticket Hypothesis
• AIPerf: Automated machine learning as an AI-HPC benchmark
• Projective Manifold Gradient Layer for Deep Rotation Regression
• Hybrid Attention for Automatic Segmentation of Whole Fetal Head in Prenatal Ultrasound Volumes
• StarEnhancer: Learning Real-Time and Style-Aware Image Enhancement
• Token Pooling in Vision Transformers
• Creating Lightweight Object Detectors with Model Compression for Deployment on Edge Devices
• S3ML: A Secure Serving System for Machine Learning Inference
• Robust, Extensible, and Fast: Teamed Classifiers for Vehicle Tracking and Vehicle Re-ID in Multi-Camera Networks
• Dual Attention MobDenseNet(DAMDNet) for Robust 3D Face Alignment
• A Survey of Large-Scale Deep Learning Serving System Optimization: Challenges and Opportunities
• Dynamic Routing Networks
• Arch-Net: Model Distillation for Architecture Agnostic Model Deployment
• Composite Binary Decomposition Networks
• TP-TIO: A Robust Thermal-Inertial Odometry with Deep ThermalPoint
• "Tom" pet robot applied to urban autism
• Information-Theoretic Understanding of Population Risk Improvement with Model Compression
• Learning degraded image classification with restoration data fidelity
• ShuffleDet: Real-Time Vehicle Detection Network in On-board Embedded UAV Imagery
• Class-dependent Compression of Deep Neural Networks
• Learning by Self-Explanation, with Application to Neural Architecture Search
• An efficient deep learning hashing neural network for mobile visual search
• Spectral Image Visualization Using Generative Adversarial Networks
• A Separable Temporal Convolution Neural Network with Attention for Small-Footprint Keyword Spotting
• ExGate: Externally Controlled Gating for Feature-based Attention in Artificial Neural Networks
• Understanding the Disharmony between Weight Normalization Family and Weight Decay: $\epsilon -$shifted $L_2$ Regularizer
• Orderly Dual-Teacher Knowledge Distillation for Lightweight Human Pose Estimation
• Complexity-aware Adaptive Training and Inference for Edge-Cloud Distributed AI Systems
• Dynamic Multi-path Neural Network
• Dense Pruning of Pointwise Convolutions in the Frequency Domain
• Optimizing the Whole-life Cost in End-to-end CNN Acceleration
• LiteDepthwiseNet: An Extreme Lightweight Network for Hyperspectral Image Classification
• Semi-tensor Product-based TensorDecomposition for Neural Network Compression
• Inter-channel Conv-TasNet for multichannel speech enhancement
• AI-based BMI Inference from Facial Images: An Application to Weight Monitoring
• Oculum afficit: Ocular Affect Recognition
• Attention-Aware Linear Depthwise Convolution for Single Image Super-Resolution
• Quantization Mimic: Towards Very Tiny CNN for Object Detection
• Dirichlet Pruning for Neural Network Compression
• FDNAS: Improving Data Privacy and Model Diversity in AutoML
• Manifestation of Image Contrast in Deep Networks
• Point-of-Care Diabetic Retinopathy Diagnosis: A Standalone Mobile Application Approach
• Semi-Streaming Architecture: A New Design Paradigm for CNN Implementation on FPGAs
• Scheduling Optimization Techniques for Neural Network Training
• Deep Learning Acceleration Techniques for Real Time Mobile Vision Applications
• Direct Federated Neural Architecture Search
• Egocentric 6-DoF Tracking of Small Handheld Objects
• Chaotic-to-Fine Clustering for Unlabeled Plant Disease Images
• Grounding Human-to-Vehicle Advice for Self-driving Vehicles
• Lightweight Mask R-CNN for Long-Range Wireless Power Transfer Systems
• Semi-Online Knowledge Distillation
• Widening and Squeezing: Towards Accurate and Efficient QNNs
• Low-memory convolutional neural networks through incremental depth-first processing
• Training a Binary Weight Object Detector by Knowledge Transfer for Autonomous Driving
• A mixed signal architecture for convolutional neural networks
• Receptive Field Broadening and Boosting for Salient Object Detection
• CompactNet: Platform-Aware Automatic Optimization for Convolutional Neural Networks
• Dynamic Slimmable Denoising Network
• RRNet: Repetition-Reduction Network for Energy Efficient Decoder of Depth Estimation
• Scale Calibrated Training: Improving Generalization of Deep Networks via Scale-Specific Normalization
• Signature-Graph Networks
• Toward Runtime-Throttleable Neural Networks
• Scalable Smartphone Cluster for Deep Learning
• EfficientPose: Efficient Human Pose Estimation with Neural Architecture Search
• Privacy Aware Person Detection in Surveillance Data
• An Efficient Quantitative Approach for Optimizing Convolutional Neural Networks
• R-TOD: Real-Time Object Detector with Minimized End-to-End Delay for Autonomous Driving
• ADDS: Adaptive Differentiable Sampling for Robust Multi-Party Learning
• MOS: A Low Latency and Lightweight Framework for Face Detection, Landmark Localization, and Head Pose Estimation
• Evaluating robustness of You Only Hear Once(YOHO) Algorithm on noisy audios in the VOICe Dataset
• MassFace: an efficient implementation using triplet loss for face recognition
• Integrating Multiple Receptive Fields through Grouped Active Convolution
• Channel Pruning via Multi-Criteria based on Weight Dependency
• LogAvgExp Provides a Principled and Performant Global Pooling Operator
• SurveilEdge: Real-time Video Query based on Collaborative Cloud-Edge Deep Learning
• Fed2: Feature-Aligned Federated Learning
• MDLdroid: a ChainSGD-reduce Approach to Mobile Deep Learning for Personal Mobile Sensing
• Learning Universal Shape Dictionary for Realtime Instance Segmentation
• The Heterogeneity Hypothesis: Finding Layer-Wise Differentiated Network Architectures
• Face Image Reflection Removal
• Regression on Deep Visual Features using Artificial Neural Networks (ANNs) to Predict Hydraulic Blockage at Culverts
• MCENET: Multi-Context Encoder Network for Homogeneous Agent Trajectory Prediction in Mixed Traffic
• ParaDiS: Parallelly Distributable Slimmable Neural Networks
• DepthwiseGANs: Fast Training Generative Adversarial Networks for Realistic Image Synthesis
• One Backward from Ten Forward, Subsampling for Large-Scale Deep Learning
• One Weight Bitwidth to Rule Them All
• Spatiotemporal Action Recognition in Restaurant Videos
• CRL: Class Representative Learning for Image Classification
• Efficient Incorporation of Multiple Latency Targets in the Once-For-All Network
• Structured Sparsification with Joint Optimization of Group Convolution and Channel Shuffle
• Realtime Rooftop Landing Site Identification and Selection in Urban City Simulation
• Resolution Switchable Networks for Runtime Efficient Image Recognition
• Toward Building Safer Smart Homes for the People with Disabilities
• CASSOD-Net: Cascaded and Separable Structures of Dilated Convolution for Embedded Vision Systems and Applications
• Embedded Systems and Computer Vision Techniques utilized in Spray Painting Robots: A Review
• Complementary Relation Contrastive Distillation
• Multi-view Feature Augmentation with Adaptive Class Activation Mapping
• Decoupled Dynamic Filter Networks
• Graph Pruning for Model Compression
• Shift-and-Balance Attention
• Go Wider: An Efficient Neural Network for Point Cloud Analysis via Group Convolutions
• Dynamic Slimmable Network
• NeuroFabric: Identifying Ideal Topologies for Training A Priori Sparse Networks
• Classifying Suspicious Content in Tor Darknet
• Deep Octonion Networks
• Similarity Transfer for Knowledge Distillation
• TASO: Time and Space Optimization for Memory-Constrained DNN Inference
• Diversifying Inference Path Selection: Moving-Mobile-Network for Landmark Recognition
• Energy Predictive Models for Convolutional Neural Networks on Mobile Platforms
• Improved Generalization of Heading Direction Estimation for Aerial Filming Using Semi-supervised Regression
• Automatic Identification and Description of Jewelry Through Computer Vision and Neural Networks for Translators and Interpreters
• Towards Inference Delivery Networks: Distributing Machine Learning with Optimality Guarantees
• BiDet: An Efficient Binarized Object Detector
• An Inter-Layer Weight Prediction and Quantization for Deep Neural Networks based on a Smoothly Varying Weight Hypothesis
• PolyResponse: A Rank-based Approach to Task-Oriented Dialogue with Application in Restaurant Search and Booking
• An Efficient Method of Training Small Models for Regression Problems with Knowledge Distillation
• Channel Pruning Guided by Classification Loss and Feature Importance
• DP-Net: Dynamic Programming Guided Deep Neural Network Compression
• Real-Time Semantic Segmentation via Auto Depth, Downsampling Joint Decision and Feature Aggregation
• Handwritten Character Recognition from Wearable Passive RFID
• Context Prior for Scene Segmentation
• Background Activation Suppression for Weakly Supervised Object Localization
• Neural Inheritance Relation Guided One-Shot Layer Assignment Search
• Prune2Edge: A Multi-Phase Pruning Pipelines to Deep Ensemble Learning in IIoT
• Context-Gated Convolution
• Simplifying Neural Networks using Formal Verification
• Fully Dynamic Inference with Deep Neural Networks
• Adaptive Test-Time Augmentation for Low-Power CPU
• StRDAN: Synthetic-to-Real Domain Adaptation Network for Vehicle Re-Identification
• FLAME: A Self-Adaptive Auto-labeling System for Heterogeneous Mobile Processors
• Cross-Channel Intragroup Sparsity Neural Network
• ADWPNAS: Architecture-Driven Weight Prediction for Neural Architecture Search
• Convolutional Neural Network for emotion recognition to assist psychiatrists and psychologists during the COVID-19 pandemic: experts opinion
• Temporally Resolution Decrement: Utilizing the Shape Consistency for Higher Computational Efficiency
• I/O Lower Bounds for Auto-tuning of Convolutions in CNNs
• APNN-TC: Accelerating Arbitrary Precision Neural Networks on Ampere GPU Tensor Cores
• Full-stack Optimization for Accelerating CNNs with FPGA Validation
• Form2Seq : A Framework for Higher-Order Form Structure Extraction
• Selective Deep Convolutional Neural Network for Low Cost Distorted Image Classification
• BRIEF: Backward Reduction of CNNs with Information Flow Analysis
• Multi-Modal Association based Grouping for Form Structure Extraction
• Elastic Neural Networks: A Scalable Framework for Embedded Computer Vision
• AutoFL: Enabling Heterogeneity-Aware Energy Efficient Federated Learning
• Leveraging Implicit Spatial Information in Global Features for Image Retrieval
• Evolving Neural Architecture Using One Shot Model
• A Survey of Machine Learning Techniques for Detecting and Diagnosing COVID-19 from Imaging
• Log-Polar Space Convolution for Convolutional Neural Networks
• Deep Learning Approximation: Zero-Shot Neural Network Speedup
• MixFaceNets: Extremely Efficient Face Recognition Networks
• Distributional Depth-Based Estimation of Object Articulation Models
• WeClick: Weakly-Supervised Video Semantic Segmentation with Click Annotations
• One-Shot Object Affordance Detection in the Wild
• Adaptive Precision Training for Resource Constrained Devices
• Learning Deep Multimodal Feature Representation with Asymmetric Multi-layer Fusion
• FOX-NAS: Fast, On-device and Explainable Neural Architecture Search
• Multi-granularity for knowledge distillation
• Sonic: A Sampling-based Online Controller for Streaming Applications
• Edge Computing Enabled by Unmanned Autonomous Vehicles
• Analyze and Design Network Architectures by Recursion Formulas
• Separable Temporal Convolution plus Temporally Pooled Attention for Lightweight High-performance Keyword Spotting
• Power-Based Attacks on Spatial DNN Accelerators
• Exploring and Improving Mobile Level Vision Transformers
• Canoe : A System for Collaborative Learning for Neural Nets
• AIP: Adversarial Iterative Pruning Based on Knowledge Transfer for Convolutional Neural Networks
• Architecture Aware Latency Constrained Sparse Neural Networks
• Does Melania Trump have a body double from the perspective of automatic face recognition?
• Tom: Leveraging trend of the observed gradients for faster convergence
• Multi-Scale Aligned Distillation for Low-Resolution Detection
• DEM Super-Resolution with EfficientNetV2
• Classification of COVID-19 from CXR Images in a 15-class Scenario: an Attempt to Avoid Bias in the System
• Summarize and Search: Learning Consensus-aware Dynamic Convolution for Co-Saliency Detection
• Approximate Random Dropout
• Internal node bagging
• Neural Architecture Search via Combinatorial Multi-Armed Bandit
• Seeing Convolution Through the Eyes of Finite Transformation Semigroup Theory: An Abstract Algebraic Interpretation of Convolutional Neural Networks
• Accelerate Distributed Stochastic Descent for Nonconvex Optimization with Momentum
• Learning to Generate Content-Aware Dynamic Detectors
• Heterogeneous Dual-Core Overlay Processor for Light-Weight CNNs
• Efficient Modelling Across Time of Human Actions and Interactions
• The Low-Resource Double Bind: An Empirical Study of Pruning for Low-Resource Machine Translation
• Robust Real-Time Pedestrian Detection on Embedded Devices
• End-to-end Keyword Spotting using Xception-1d
• Space-Time-Separable Graph Convolutional Network for Pose Forecasting
• Online Filter Clustering and Pruning for Efficient Convnets
• Weight Evolution: Improving Deep Neural Networks Training through Evolving Inferior Weight Values
• Super Interaction Neural Network
• Sub-bit Neural Networks: Learning to Compress and Accelerate Binary Neural Networks
• DAIL: Dataset-Aware and Invariant Learning for Face Recognition
• Accelerate 3D Object Processing via Spectral Layout
• Modular network for high accuracy object detection
• A Distributed Framework to Orchestrate Video Analytics Applications
• Compressing Facial Makeup Transfer Networks by Collaborative Distillation and Kernel Decomposition
• Non-contact Real time Eye Gaze Mapping System Based on Deep Convolutional Neural Network
• Low-Rank+Sparse Tensor Compression for Neural Networks
• Compression of descriptor models for mobile applications
• Learning to Prevent Leakage: Privacy-Preserving Inference in the Mobile Cloud
• Deep Model Compression Via Two-Stage Deep Reinforcement Learning
• Object Detection-Based Variable Quantization Processing
• Virtual Experience to Real World Application: Sidewalk Obstacle Avoidance Using Reinforcement Learning for Visually Impaired
• Artificial Intelligence in Surgery: Neural Networks and Deep Learning
• Multi-Precision Quantized Neural Networks via Encoding Decomposition of -1 and +1
• Self-grouping Convolutional Neural Networks
• BAMSProd: A Step towards Generalizing the Adaptive Optimization Methods to Deep Binary Model
• Communication-Efficient Separable Neural Network for Distributed Inference on Edge Devices
• Bosch Deep Learning Hardware Benchmark
• Normalized Label Distribution: Towards Learning Calibrated, Adaptable and Efficient Activation Maps
• Mobile Networks for Computer Go
• Machine Vision for Improved Human-Robot Cooperation in Adverse Underwater Conditions
• Compressing Deep Convolutional Neural Networks by Stacking Low-dimensional Binary Convolution Filters
• Scalability vs. Utility: Do We Have to Sacrifice One for the Other in Data Importance Quantification?
• Emergent symbolic language based deep medical image classification
• Incremental Cross-Domain Adaptation for Robust Retinopathy Screening via Bayesian Deep Learning
• Transfer Learning with Binary Neural Networks
• Learning Connectivity of Neural Networks from a Topological Perspective
• Object Detection in the Context of Mobile Augmented Reality
• Explainable, automated urban interventions to improve pedestrian and vehicle safety
• Multilayer Dense Connections for Hierarchical Concept Classification
• GPCA: A Probabilistic Framework for Gaussian Process Embedded Channel Attention
• Towards Modality Transferable Visual Information Representation with Optimal Model Compression
• IC-Network: Efficient Structure for Convolutional Neural Networks
• ERNet Family: Hardware-Oriented CNN Models for Computational Imaging Using Block-Based Inference
• Pacemaker: Intermediate Teacher Knowledge Distillation For On-The-Fly Convolutional Neural Network
• A Unifying Framework of Bilinear LSTMs
• Phantom: A High-Performance Computational Core for Sparse Convolutional Neural Networks
• A Benchmarking Framework for Interactive 3D Applications in the Cloud
• Attention-Guided Lightweight Network for Real-Time Segmentation of Robotic Surgical Instruments
• Boundary-Aware Dense Feature Indicator for Single-Stage 3D Object Detection from Point Clouds
• Accelerating Deep Learning Inference with Cross-Layer Data Reuse on GPUs
• DD-CNN: Depthwise Disout Convolutional Neural Network for Low-complexity Acoustic Scene Classification
• Fast Portrait Segmentation with Highly Light-weight Network
• Tamed Warping Network for High-Resolution Semantic Video Segmentation
• Adaptive and Iteratively Improving Recurrent Lateral Connections
• Unknown Identity Rejection Loss: Utilizing Unlabeled Data for Face Recognition
• On-device Filtering of Social Media Images for Efficient Storage
• One of these (Few) Things is Not Like the Others
• Comb Convolution for Efficient Convolutional Architecture
• All-Weather Object Recognition Using Radar and Infrared Sensing
• Resource-Efficient Speech Mask Estimation for Multi-Channel Speech Enhancement
• RethNet: Object-by-Object Learning for Detecting Facial Skin Problems
• NeuroMAX: A High Throughput, Multi-Threaded, Log-Based Accelerator for Convolutional Neural Networks
• JNR: Joint-based Neural Rig Representation for Compact 3D Face Modeling
• Novel Adaptive Binary Search Strategy-First Hybrid Pyramid- and Clustering-Based CNN Filter Pruning Method without Parameters Setting
• POD: Practical Object Detection with Scale-Sensitive Network
• Architecture Agnostic Neural Networks
• What Happens on the Edge, Stays on the Edge: Toward Compressive Deep Learning
• Screen Tracking for Clinical Translation of Live Ultrasound Image Analysis Methods
• Conditioned Time-Dilated Convolutions for Sound Event Detection
• Deeply Shared Filter Bases for Parameter-Efficient Convolutional Neural Networks
• Universal Physical Camouflage Attacks on Object Detectors
• Right for the Right Reason: Making Image Classification Robust
• Cyclic orthogonal convolutions for long-range integration of features
• Discretization-Aware Architecture Search
• Mobile Recognition of Wikipedia Featured Sites using Deep Learning and Crowd-sourced Imagery
• MERIT: Tensor Transform for Memory-Efficient Vision Processing on Parallel Architectures
• Joslim: Joint Widths and Weights Optimization for Slimmable Neural Networks
• Recurrent Residual Module for Fast Inference in Videos
• DupNet: Towards Very Tiny Quantized CNN with Improved Accuracy for Face Detection
• BNAS-v2: Memory-efficient and Performance-collapse-prevented Broad Neural Architecture Search
• MSR-DARTS: Minimum Stable Rank of Differentiable Architecture Search
• User-centric Composable Services: A New Generation of Personal Data Analytics
• Automatic Photo to Ideophone Manga Matching
• JIT-Masker: Efficient Online Distillation for Background Matting
• Grow-Push-Prune: aligning deep discriminants for effective structural network compression
• Sparsifying and Down-scaling Networks to Increase Robustness to Distortions
• Model Adaption Object Detection System for Robot
• AutoBSS: An Efficient Algorithm for Block Stacking Style Search
• A Comparative Study of U-Net Topologies for Background Removal in Histopathology Images
• LETI: Latency Estimation Tool and Investigation of Neural Networks inference on Mobile GPU
• Activation Map Adaptation for Effective Knowledge Distillation
• Architecture-aware Network Pruning for Vision Quality Applications
• A Study on Trees's Knots Prediction from their Bark Outer-Shape
• Be Your Own Best Competitor! Multi-Branched Adversarial Knowledge Transfer
• Lightweight, Dynamic Graph Convolutional Networks for AMR-to-Text Generation
• EIS -- a family of activation functions combining Exponential, ISRU, and Softplus
• Approximation Algorithms for Cascading Prediction Models
• Accelerating Neural Network Inference by Overflow Aware Quantization
• Radius Adaptive Convolutional Neural Network
• Coarse and fine-grained automatic cropping deep convolutional neural network
• Softer Pruning, Incremental Regularization
• Robustness-aware 2-bit quantization with real-time performance for neural network
• CooGAN: A Memory-Efficient Framework for High-Resolution Facial Attribute Editing
• Optimising the Performance of Convolutional Neural Networks across Computing Systems using Transfer Learning
• Large Scale Indexing of Generic Medical Image Data using Unbiased Shallow Keypoints and Deep CNN Features
• Dense Dual-Path Network for Real-time Semantic Segmentation
• Effective Model Compression via Stage-wise Pruning
• A Comparative Study of High-Recall Real-Time Semantic Segmentation Based on Swift Factorized Network
• MixMix: All You Need for Data-Free Compression Are Feature and Data Mixing
• Feature Statistics Guided Efficient Filter Pruning
• Cross-filter compression for CNN inference acceleration
• MicroNet for Efficient Language Modeling
• DSXplore: Optimizing Convolutional Neural Networks via Sliding-Channel Convolutions
• GHFP: Gradually Hard Filter Pruning
• RetinotopicNet: An Iterative Attention Mechanism Using Local Descriptors with Global Context
• Deep Medical Image Analysis with Representation Learning and Neuromorphic Computing
• SmartDeal: Re-Modeling Deep Network Weights for Efficient Inference and Training
• Block-Cyclic Stochastic Coordinate Descent for Deep Neural Networks
• GraCIAS: Grassmannian of Corrupted Images for Adversarial Security
• Towards Building a Real Time Mobile Device Bird Counting System Through Synthetic Data Training and Model Compression
• Classification of Industrial Control Systems screenshots using Transfer Learning
• CRCEN: A Generalized Cost-sensitive Neural Network Approach for Imbalanced Classification
• Hands-on Guidance for Distilling Object Detectors
• Spectral Tensor Train Parameterization of Deep Learning Layers
• Selective Output Smoothing Regularization: Regularize Neural Networks by Softening Output Distributions
• Unpaired Single-Image Depth Synthesis with cycle-consistent Wasserstein GANs
• An Analysis of Object Representations in Deep Visual Trackers
• Network Space Search for Pareto-Efficient Spaces
• FSD: Feature Skyscraper Detector for Stem End and Blossom End of Navel Orange
• Landmark-Aware and Part-based Ensemble Transfer Learning Network for Facial Expression Recognition from Static images
• Carrying out CNN Channel Pruning in a White Box
• Towards Real-Time DNN Inference on Mobile Platforms with Model Pruning and Compiler Optimization
• Effects of annotation granularity in deep learning models for histopathological images
• SafeNet: An Assistive Solution to Assess Incoming Threats for Premises
• Are Accelerometers for Activity Recognition a Dead-end?
• Analyzing the Dependency of ConvNets on Spatial Information
• PolyScientist: Automatic Loop Transformations Combined with Microkernels for Optimization of Deep Learning Primitives
• Deep Learning Derived Histopathology Image Score for Increasing Phase 3 Clinical Trial Probability of Success
• CheckNet: Secure Inference on Untrusted Devices
• Boosted Locality Sensitive Hashing: Discriminative Binary Codes for Source Separation
• Diagonal Memory Optimisation for Machine Learning on Micro-controllers
• Spatially Attentive Output Layer for Image Classification
• Width Transfer: On the (In)variance of Width Optimization
• Explore the Knowledge contained in Network Weights to Obtain Sparse Neural Networks
• Group Feature Learning and Domain Adversarial Neural Network for aMCI Diagnosis System Based on EEG
• Itsy Bitsy SpiderNet: Fully Connected Residual Network for Fraud Detection
• SpotPatch: Parameter-Efficient Transfer Learning for Mobile Object Detection
• Anchor-based Plain Net for Mobile Image Super-Resolution
• Improve SGD Training via Aligning Mini-batches
• Extreme Value Preserving Networks
• Automatic Non-Linear Video Editing Transfer
• Balancing Accuracy and Latency in Multipath Neural Networks
• Distilling Image Classifiers in Object Detectors
• Place recognition survey: An update on deep learning approaches
• Abiotic Stress Prediction from RGB-T Images of Banana Plantlets
• Deep Tiny Network for Recognition-Oriented Face Image Quality Assessment
• Continuous Trade-off Optimization between Fast and Accurate Deep Face Detectors
• DAC: Data-free Automatic Acceleration of Convolutional Networks
• RingCNN: Exploiting Algebraically-Sparse Ring Tensors for Energy-Efficient CNN-Based Computational Imaging
• Filter Pre-Pruning for Improved Fine-tuning of Quantized Deep Neural Networks
• LightFuse: Lightweight CNN based Dual-exposure Fusion
• Detection of preventable fetal distress during labor from scanned cardiotocogram tracings using deep learning
• AsymmNet: Towards ultralight convolution neural networks using asymmetrical bottlenecks
• Handcrafted vs Deep Learning Classification for Scalable Video QoE Modeling
• AffectiveNet: Affective-Motion Feature Learningfor Micro Expression Recognition
• VisBuddy -- A Smart Wearable Assistant for the Visually Challenged
• Visual Domain Adaptation for Monocular Depth Estimation on Resource-Constrained Hardware
• On Generalization of Adaptive Methods for Over-parameterized Linear Regression
• Annealing Knowledge Distillation
• HPTQ: Hardware-Friendly Post Training Quantization
• ISyNet: Convolutional Neural Networks design for AI accelerator
• Prioritized Subnet Sampling for Resource-Adaptive Supernet Training
• Energy Efficient Hardware for On-Device CNN Inference via Transfer Learning
• Integrating Information Theory and Adversarial Learning for Cross-modal Retrieval
• Binary Neural Network for Speaker Verification
• ConAM: Confidence Attention Module for Convolutional Neural Networks
• The Efficiency Misnomer
• Efficient Video Understanding via Layered Multi Frame-Rate Analysis
• Machine Learning with Clos Networks
• Training Multi-bit Quantized and Binarized Networks with A Learnable Symmetric Quantizer
• High-quality Low-dose CT Reconstruction Using Convolutional Neural Networks with Spatial and Channel Squeeze and Excitation
• Training Domain Specific Models for Energy-Efficient Object Detection
• How Well Do Sparse Imagenet Models Transfer?
• Learning Pruned Structure and Weights Simultaneously from Scratch: an Attention based Approach
• A Lightweight Graph Transformer Network for Human Mesh Reconstruction from 2D Human Pose
• Multi-Glimpse Network: A Robust and Efficient Classification Architecture based on Recurrent Downsampled Attention
• Dataset Culling: Towards Efficient Training Of Distillation-Based Domain Specific Models
• Convergence and Stability of the Stochastic Proximal Point Algorithm with Momentum
• Small-Group Learning, with Application to Neural Architecture Search
• A Survey on Green Deep Learning
• MICIK: MIning Cross-Layer Inherent Similarity Knowledge for Deep Model Compression
• Expert Human-Level Driving in Gran Turismo Sport Using Deep Reinforcement Learning with Image-based Representation
• Depth estimation on embedded computers for robot swarms in forest
• Efficient Structured Pruning and Architecture Searching for Group Convolution
• Factorial Convolution Neural Networks
• Searching for TrioNet: Combining Convolution with Local and Global Self-Attention
• Knowledge Distillation By Sparse Representation Matching
• VideoPose: Estimating 6D object pose from videos
• Delving into Rectifiers in Style-Based Image Translation
• Illumination-invariant Face recognition by fusing thermal and visual images via gradient transfer
• A Dense Tensor Accelerator with Data Exchange Mesh for DNN and Vision Workloads
• Cluster Regularized Quantization for Deep Networks Compression
• A Generalized Meta-loss function for regression and classification using privileged information
• Asian Giant Hornet Control based on Image Processing and Biological Dispersal
• Improving Differentiable Architecture Search with a Generative Model
• Real-time Action Recognition with Dissimilarity-based Training of Specialized Module Networks
• Learning from Mistakes based on Class Weighting with Application to Neural Architecture Search
• Real-time Human Detection Model for Edge Devices
• Enabling Design Methodologies and Future Trends for Edge AI: Specialization and Co-design
• Penetrating the Fog: the Path to Efficient CNN Models
• Two-Stage Monte Carlo Denoising with Adaptive Sampling and Kernel Pool
• Differentiable Network Adaption with Elastic Search Space
• Dynamic Domain Adaptation for Efficient Inference
• Elastic Neural Networks for Classification
• Transfer Learning-based Real-time Handgun Detection
• clcNet: Improving the Efficiency of Convolutional Neural Network using Channel Local Convolutions
• BitSplit-Net: Multi-bit Deep Neural Network with Bitwise Activation Function
• On evaluating CNN representations for low resource medical image classification
• Parallel Blockwise Knowledge Distillation for Deep Neural Network Compression
• MyFood: A Food Segmentation and Classification System to Aid Nutritional Monitoring
• BlockDoor: Blocking Backdoor Based Watermarks in Deep Neural Networks
• MobileFace: 3D Face Reconstruction with Efficient CNN Regression
• Improved knowledge distillation by utilizing backward pass knowledge in neural networks
• Empirical Study on the Software Engineering Practices in Open Source ML Package Repositories
• Putting 3D Spatially Sparse Networks on a Diet
• PokeBNN: A Binary Pursuit of Lightweight Accuracy
• Efficient Fusion of Sparse and Complementary Convolutions
• 3rd Place Solution for NeurIPS 2021 Shifts Challenge: Vehicle Motion Prediction
• A CNN Accelerator on FPGA Using Depthwise Separable Convolution
• Green Accelerated Hoeffding Tree
• Hybrid Cosine Based Convolutional Neural Networks
• Semantic Relation Preserving Knowledge Distillation for Image-to-Image Translation
• Towards a Sample Efficient Reinforcement Learning Pipeline for Vision Based Robotics
• C2S2: Cost-aware Channel Sparse Selection for Progressive Network Pruning
• Content-adaptive Representation Learning for Fast Image Super-resolution
• MBS: Macroblock Scaling for CNN Model Reduction
• DTNN: Energy-efficient Inference with Dendrite Tree Inspired Neural Networks for Edge Vision Applications
• PSRR-MaxpoolNMS: Pyramid Shifted MaxpoolNMS with Relationship Recovery
• Scorpion detection and classification systems based on computer vision and deep learning for health security purposes
• Action Recognition with Kernel-based Graph Convolutional Networks
• Single Image Depth Prediction with Wavelet Decomposition
• CDN-MEDAL: Two-stage Density and Difference Approximation Framework for Motion Analysis
• Deep Virtual Networks for Memory Efficient Inference of Multiple Tasks
• LLC: Accurate, Multi-purpose Learnt Low-dimensional Binary Codes
• Backtracking gradient descent method for general $C^1$ functions, with applications to Deep Learning
• Lightweight Combinational Machine Learning Algorithm for Sorting Canine Torso Radiographs
• Go Small and Similar: A Simple Output Decay Brings Better Performance
• SGE net: Video object detection with squeezed GRU and information entropy map
• Depthwise Separable Convolutions Allow for Fast and Memory-Efficient Spectral Normalization
• using multiple losses for accurate facial age estimation
• Quantized Neural Networks via {-1, +1} Encoding Decomposition and Acceleration
• Self-supervised Knowledge Distillation Using Singular Value Decomposition
• Gradient-Based Interpretability Methods and Binarized Neural Networks
• CoEdge: Cooperative DNN Inference with Adaptive Workload Partitioning over Heterogeneous Edge Devices
• LV-BERT: Exploiting Layer Variety for BERT
• Auto Deep Compression by Reinforcement Learning Based Actor-Critic Structure
• Doing good by fighting fraud: Ethical anti-fraud systems for mobile payments
• $S^2$-LBI: Stochastic Split Linearized Bregman Iterations for Parsimonious Deep Learning
• Mutually-aware Sub-Graphs Differentiable Architecture Search
• Performance landscape of resource-constrained platforms targeting DNNs
• Wearable Travel Aid for Environment Perception and Navigation of Visually Impaired People
• CompConv: A Compact Convolution Module for Efficient Feature Learning