Add to Chrome
Add to Firefox
Add to EdgeECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks
Oct 08, 2019
Channel attention has recently demonstrated to offer great potential in improving the performance of deep convolutional neural networks (CNNs). However, most existing methods dedicate to developing more sophisticated attention modules to achieve better performance, inevitably increasing the computational burden. To overcome the paradox of performance and complexity trade-off, this paper makes an attempt to investigate an extremely lightweight attention module for boosting the performance of deep CNNs. In particular, we propose an Efficient Channel Attention (ECA) module, which only involves $k (k < 9)$ parameters but brings clear performance gain. By revisiting the channel attention module in SENet, we empirically show avoiding dimensionality reduction and appropriate cross-channel interaction are important to learn effective channel attention. Therefore, we propose a local cross-channel interaction strategy without dimension reduction, which can be efficiently implemented by a fast 1D convolution. Furthermore, we develop a function of channel dimension to adaptively determine kernel size of 1D convolution, which stands for coverage of local cross-channel interaction. Our ECA module can be flexibly incorporated into existing CNN architectures, and the resulting CNNs are named by ECA-Net. We extensively evaluate the proposed ECA-Net on image classification, object detection and instance segmentation with backbones of ResNets and MobileNetV2. The experimental results show our ECA-Net is more efficient while performing favorably against its counterparts. The source code and models can be available at https://github.com/BangguWu/ECANet.
