ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
Xiangyu Zhang,Xinyu Zhou,Mengxiao Lin,Jian Sun +3 more
- pp 6848-6856
Reads0
Chats0
TLDR
ShuffleNet as discussed by the authors utilizes two new operations, pointwise group convolution and channel shuffle, to greatly reduce computation cost while maintaining accuracy, and achieves an actual speedup over AlexNet while maintaining comparable accuracy.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 [12] on ImageNet classification task, under the computation budget of 40 MFLOPs. On an ARM-based mobile device, ShuffleNet achieves ~13A— actual speedup over AlexNet while maintaining comparable accuracy.read more
Citations
More filters
Journal ArticleDOI
Unified Diagnosis Framework for Automated Nuclear Cataract Grading Based on Smartphone Slit-Lamp Images
TL;DR: The proposed validation method could evaluate a cataract severity in 29 ms and the entire classification process in less than 1s and can improve the accuracy of the examination, reduce misdiagnosis rate and the difficulty of the doctor’s examination.
Journal ArticleDOI
Learning to Recognize Chest-Xray Images Faster and More Efficiently Based on Multi-Kernel Depthwise Convolution
TL;DR: This paper proposes the multi-kernel depthwise convolution (MD-Conv) which contains depthwiseconvolution kernels with different filter sizes in one depthwise Convolution layer which has high calculation efficiency and few parameters and can be easily applied in modern lightweight networks to replace the normal depth Wise convolution layer.
Journal ArticleDOI
Energy-Efficient Real-Time UAV Object Detection on Embedded Platforms
TL;DR: This article proposes an energy-efficient system for real-time UAV object detection on an embedded platform that can achieve speed of 28.5 FPS and 2.7-FPS/W energy efficiency on the data set from 2018 low-power object detection challenges (LPODCs).
Journal ArticleDOI
TIB-Net: Drone Detection Network With Tiny Iterative Backbone
TL;DR: This work develops a drone detection network with tiny iterative backbone named TIB-Net, and proposes a structure called cyclic pathway, which enhances the capability to extract effective features of small object, and integrates it into existing efficient method Extremely Tiny Face Detector (EXTD).
Posted Content
Reinforced Evolutionary Neural Architecture Search.
TL;DR: The Reinforced Evolutionary Neural Architecture Search (RE- NAS) is proposed, which is an evolutionary method with the reinforced mutation for NAS that achieves a competitive result on CIFAR-10 and obtains a powerful network architecture, RENASNet.
References
More filters
Proceedings ArticleDOI
Deep Residual Learning for Image Recognition
TL;DR: In this article, the authors proposed a residual learning framework to ease the training of networks that are substantially deeper than those used previously, which won the 1st place on the ILSVRC 2015 classification task.
Proceedings Article
ImageNet Classification with Deep Convolutional Neural Networks
TL;DR: The state-of-the-art performance of CNNs was achieved by Deep Convolutional Neural Networks (DCNNs) as discussed by the authors, which consists of five convolutional layers, some of which are followed by max-pooling layers, and three fully-connected layers with a final 1000-way softmax.
Proceedings Article
Very Deep Convolutional Networks for Large-Scale Image Recognition
Karen Simonyan,Andrew Zisserman +1 more
TL;DR: This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers.
Proceedings Article
Very Deep Convolutional Networks for Large-Scale Image Recognition
Karen Simonyan,Andrew Zisserman +1 more
TL;DR: In this paper, the authors investigated the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting and showed that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 layers.
Proceedings ArticleDOI
ImageNet: A large-scale hierarchical image database
TL;DR: A new database called “ImageNet” is introduced, a large-scale ontology of images built upon the backbone of the WordNet structure, much larger in scale and diversity and much more accurate than the current image datasets.