DenseNet: Implementing Efficient ConvNet Descriptor Pyramids

Abstract

Convolutional Neural Networks (CNNs) can provide accurate object classification. They can be extended to perform object detection by iterating over dense or selected proposed object regions. However, the runtime of such detectors scales as the total number and/or area of regions to examine per image, and training such detectors may be prohibitively slow. However, for some CNN classifier topologies, it is possible to share significant work among overlapping regions to be classified. This paper presents DenseNet, an open source system that computes dense, multiscale features from the convolutional layers of a CNN based object classifier. Future work will involve training efficient object detectors with DenseNet feature descriptors.

References in corpus (2)

• DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition
• Fast Image Scanning with Deep Max-Pooling Convolutional Neural Networks

Cited by in corpus (83)

• Richer Convolutional Features for Edge Detection
• Particular object retrieval with integral max-pooling of CNN activations
• Speeding up Convolutional Neural Networks with Low Rank Expansions
• A Survey of Convolutional Neural Networks: Analysis, Applications, and Prospects
• Pneumonia Detection on chest X-ray images Using Ensemble of Deep Convolutional Neural Networks
• ImageBERT: Cross-modal Pre-training with Large-scale Weak-supervised Image-Text Data
• Clinically Accurate Chest X-Ray Report Generation
• Recent Advances in Object Detection in the Age of Deep Convolutional Neural Networks
• Aff-Wild2: Extending the Aff-Wild Database for Affect Recognition
• Terminal Brain Damage: Exposing the Graceless Degradation in Deep Neural Networks Under Hardware Fault Attacks
• ChineseFoodNet: A large-scale Image Dataset for Chinese Food Recognition
• Pixel-wise Deep Learning for Contour Detection
• Deformable Part Models are Convolutional Neural Networks
• Hi Sigma, do I have the Coronavirus?: Call for a New Artificial Intelligence Approach to Support Health Care Professionals Dealing With The COVID-19 Pandemic
• Beyond saliency: understanding convolutional neural networks from saliency prediction on layer-wise relevance propagation
• Size Matters
• Automated Human Cell Classification in Sparse Datasets using Few-Shot Learning
• Deep Neural Network Fingerprinting by Conferrable Adversarial Examples
• Compact Convolutional Neural Network Cascade for Face Detection
• 3D-SIS: 3D Semantic Instance Segmentation of RGB-D Scans
• Attended Temperature Scaling: A Practical Approach for Calibrating Deep Neural Networks
• Where to Focus: Query Adaptive Matching for Instance Retrieval Using Convolutional Feature Maps
• COVID-19-CT-CXR: a freely accessible and weakly labeled chest X-ray and CT image collection on COVID-19 from biomedical literature
• Exploring the Design Space of Deep Convolutional Neural Networks at Large Scale
• DeepEdge: A Multi-Scale Bifurcated Deep Network for Top-Down Contour Detection
• A Fast Face Detection Method via Convolutional Neural Network
• Supervised classification of Dermatological diseases by Deep learning
• Efficient Deep Neural Networks
• Towards More Transparent and Accurate Cancer Diagnosis with an Unsupervised CAE Approach
• Generative Adversarial Networks with Decoder-Encoder Output Noise
• Adversarial Robustness Assessment: Why both $L_0$ and $L_{\infty }$ Attacks Are Necessary
• FDFtNet: Facing Off Fake Images using Fake Detection Fine-tuning Network
• Rapid Exact Signal Scanning with Deep Convolutional Neural Networks
• Multi-scale Orderless Pooling of Deep Convolutional Activation Features
• SALR: Sharpness-aware Learning Rate Scheduler for Improved Generalization
• CheXseg: Combining Expert Annotations with DNN-generated Saliency Maps for X-ray Segmentation
• Towards Interpretable Face Recognition
• Compressing Deep Neural Networks via Layer Fusion
• No One Representation to Rule Them All: Overlapping Features of Training Methods
• Contour Detection Using Cost-Sensitive Convolutional Neural Networks
• QuTiBench: Benchmarking Neural Networks on Heterogeneous Hardware
• RepNAS: Searching for Efficient Re-parameterizing Blocks
• Enhancing a Neurocognitive Shared Visuomotor Model for Object Identification, Localization, and Grasping With Learning From Auxiliary Tasks
• Learning Prototype-oriented Set Representations for Meta-Learning
• MosAIc: Finding Artistic Connections across Culture with Conditional Image Retrieval
• Kernelized Deep Convolutional Neural Network for Describing Complex Images
• Self-supervised Model Based on Masked Autoencoders Advance CT Scans Classification
• Tangent Space Separability in Feedforward Neural Networks
• Towards Trainable Saliency Maps in Medical Imaging
• Unsupervised Temperature Scaling: An Unsupervised Post-Processing Calibration Method of Deep Networks
• Plexus Convolutional Neural Network (PlexusNet): A novel neural network architecture for histologic image analysis
• Symbolic Semantic Segmentation and Interpretation of COVID-19 Lung Infections in Chest CT volumes based on Emergent Languages
• Ptolemy: Architecture Support for Robust Deep Learning
• Auditing AI models for Verified Deployment under Semantic Specifications
• Consensus-Driven Propagation in Massive Unlabeled Data for Face Recognition
• Genetic Network Architecture Search
• Skeleton-based Hand-Gesture Recognition with Lightweight Graph Convolutional Networks
• Extending Label Smoothing Regularization with Self-Knowledge Distillation
• Supervised dimensionality reduction by a Linear Discriminant Analysis on pre-trained CNN features
• Finding New Diagnostic Information for Detecting Glaucoma using Neural Networks
• Automatic Labeled LiDAR Data Generation based on Precise Human Model
• Multi-label Pixelwise Classification for Reconstruction of Large-scale Urban Areas
• PoseConvGRU: A Monocular Approach for Visual Ego-motion Estimation by Learning
• Deep Learning for Prostate Pathology
• Nonlinear Collaborative Scheme for Deep Neural Networks
• 0.71-Å resolution electron tomography enabled by deep learning aided information recovery
• Convolutional Channel Features
• Classification of Long Noncoding RNA Elements Using Deep Convolutional Neural Networks and Siamese Networks
• EDEN: Enabling Energy-Efficient, High-Performance Deep Neural Network Inference Using Approximate DRAM
• End-to-end training of deep kernel map networks for image classification
• An Intelligent Remote Sensing Image Quality Inspection System
• A Fast Deep Learning Network for Automatic Image Auto-Straightening
• Pathology-Aware Generative Adversarial Networks for Medical Image Augmentation
• Fast Generation of High Fidelity RGB-D Images by Deep-Learning with Adaptive Convolution
• Identification of complex mixtures for Raman spectroscopy using a novel scheme based on a new multi-label deep neural network
• Diagonal Memory Optimisation for Machine Learning on Micro-controllers
• Past, Present, and Future Approaches Using Computer Vision for Animal Re-Identification from Camera Trap Data
• Unique Identification of Macaques for Population Monitoring and Control
• Deep Learning of Human Perception in Audio Event Classification
• Dense Forecasting of Wildfire Smoke Particulate Matter Using Sparsity Invariant Convolutional Neural Networks
• A Novel Unsupervised Post-Processing Calibration Method for DNNS with Robustness to Domain Shift
• Speeding up convolutional networks pruning with coarse ranking
• A Survey on Green Deep Learning