ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices

Abstract

We introduce an extremely computation-efficient CNN architecture named ShuffleNet, which is designed specially for mobile devices with very limited computing power (e.g., 10-150 MFLOPs). The new architecture utilizes two new operations, pointwise group convolution and channel shuffle, to greatly reduce computation cost while maintaining accuracy. Experiments on ImageNet classification and MS COCO object detection demonstrate the superior performance of ShuffleNet over other structures, e.g. lower top-1 error (absolute 7.8%) than recent MobileNet on ImageNet classification task, under the computation budget of 40 MFLOPs. On an ARM-based mobile device, ShuffleNet achieves ~13x actual speedup over AlexNet while maintaining comparable accuracy.

References in corpus (26)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Distilling the Knowledge in a Neural Network
• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• Caffe: Convolutional Architecture for Fast Feature Embedding
• Fully Convolutional Networks for Semantic Segmentation
• Squeeze-and-Excitation Networks
• DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradients
• Going Deeper with Convolutions
• Searching for Activation Functions
• Rethinking the Inception Architecture for Computer Vision
• Speeding up Convolutional Neural Networks with Low Rank Expansions
• Incremental Network Quantization: Towards Lossless CNNs with Low-Precision Weights
• Learning Structured Sparsity in Deep Neural Networks
• Aggregated Residual Transformations for Deep Neural Networks
• Xception: Deep Learning with Depthwise Separable Convolutions
• Speeding-up Convolutional Neural Networks Using Fine-tuned CP-Decomposition
• Fast Training of Convolutional Networks through FFTs
• Fast Convolutional Nets With fbfft: A GPU Performance Evaluation
• Flattened Convolutional Neural Networks for Feedforward Acceleration
• PVANET: Deep but Lightweight Neural Networks for Real-time Object Detection
• Show and Tell: A Neural Image Caption Generator
• Interleaved Group Convolutions for Deep Neural Networks
• Design of Efficient Convolutional Layers using Single Intra-channel Convolution, Topological Subdivisioning and Spatial "Bottleneck" Structure
• LCNN: Lookup-based Convolutional Neural Network

Cited by in corpus (40)

• Knowledge Distillation: A Survey
• Deep Face Recognition: A Survey
• Advancements in Image Classification using Convolutional Neural Network
• Rosetta: Large scale system for text detection and recognition in images
• Pruning by Explaining: A Novel Criterion for Deep Neural Network Pruning
• ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design
• Image Forgery Localization Based on Multi-Scale Convolutional Neural Networks
• Insect pest image detection and recognition based on bio-inspired methods
• FastDeepIoT: Towards Understanding and Optimizing Neural Network Execution Time on Mobile and Embedded Devices
• PEPSI++: Fast and Lightweight Network for Image Inpainting
• Review of data analysis in vision inspection of power lines with an in-depth discussion of deep learning technology
• Real-time monitoring of driver drowsiness on mobile platforms using 3D neural networks
• Environmental Sound Classification on the Edge: A Pipeline for Deep Acoustic Networks on Extremely Resource-Constrained Devices
• Parameter-Efficient Person Re-identification in the 3D Space
• BNAS:An Efficient Neural Architecture Search Approach Using Broad Scalable Architecture
• MobileFaceNets: Efficient CNNs for Accurate Real-Time Face Verification on Mobile Devices
• Improving Efficiency in Convolutional Neural Network with Multilinear Filters
• Medical Image Classification Using Transfer Learning and Chaos Game Optimization on the Internet of Medical Things
• TinyML for Ubiquitous Edge AI
• Size Matters
• RefineDetLite: A Lightweight One-stage Object Detection Framework for CPU-only Devices
• Channel Compression: Rethinking Information Redundancy among Channels in CNN Architecture
• Sequential Image-based Attention Network for Inferring Force Estimation without Haptic Sensor
• Evolutionary-Neural Hybrid Agents for Architecture Search
• LeanConvNets: Low-cost Yet Effective Convolutional Neural Networks
• Improving Rotated Text Detection with Rotation Region Proposal Networks
• ThreshNet: An Efficient DenseNet Using Threshold Mechanism to Reduce Connections
• Neural Network Methods for Radiation Detectors and Imaging
• Making EfficientNet More Efficient: Exploring Batch-Independent Normalization, Group Convolutions and Reduced Resolution Training
• HSD-CNN: Hierarchically self decomposing CNN architecture using class specific filter sensitivity analysis
• Evolution under Length Constraints for CNN Architecture design
• Learning towards Minimum Hyperspherical Energy
• FrostNet: Towards Quantization-Aware Network Architecture Search
• ACDC: Weight Sharing in Atom-Coefficient Decomposed Convolution
• Balanced Binary Neural Networks with Gated Residual
• LIGAR: Lightweight General-purpose Action Recognition
• ShuffleDet: Real-Time Vehicle Detection Network in On-board Embedded UAV Imagery
• PSDNet and DPDNet: Efficient channel expansion, Depthwise-Pointwise-Depthwise Inverted Bottleneck Block
• Auto Deep Compression by Reinforcement Learning Based Actor-Critic Structure
• Efficient Structured Pruning and Architecture Searching for Group Convolution