ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
This work addresses the need for efficient neural networks for mobile applications, representing a novel method for a known bottleneck rather than an incremental improvement.
The paper tackled the problem of designing efficient convolutional neural networks for mobile devices with limited computing power, achieving a 7.8% lower top-1 error than MobileNet on ImageNet under a 40 MFLOPs budget and ~13x speedup over AlexNet on an ARM-based device while maintaining accuracy.
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.