ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks

Recently, channel attention mechanism has 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 for achieving better performance, which inevitably increase model complexity. To overcome the paradox of performance and complexity trade-off, this paper proposes an Efficient Channel Attention (ECA) module, which only involves a handful of parameters while bringing clear performance gain. By dissecting the channel attention module in SENet, we empirically show avoiding dimensionality reduction is important for learning channel attention, and appropriate cross-channel interaction can preserve performance while significantly decreasing model complexity. Therefore, we propose a local cross-channel interaction strategy without dimensionality reduction, which can be efficiently implemented via $1D$ convolution. Furthermore, we develop a method to adaptively select kernel size of $1D$ convolution, determining coverage of local cross-channel interaction. The proposed ECA module is efficient yet effective, e.g., the parameters and computations of our modules against backbone of ResNet50 are 80 vs. 24.37M and 4.7e-4 GFLOPs vs. 3.86 GFLOPs, respectively, and the performance boost is more than 2% in terms of Top-1 accuracy. We extensively evaluate our ECA module on image classification, object detection and instance segmentation with backbones of ResNets and MobileNetV2. The experimental results show our module is more efficient while performing favorably against its counterparts.

PDF Abstract CVPR 2020 PDF CVPR 2020 Abstract

Datasets


Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Image Classification ImageNet ECA-Net (ResNet-152) Top 1 Accuracy 78.92% # 427
Top 5 Accuracy 94.55% # 134
Number of params 57.40M # 144
Image Classification ImageNet ECA-Net (ResNet-50) Top 1 Accuracy 77.48% # 475
Top 5 Accuracy 93.68% # 161
Number of params 24.37M # 238
Image Classification ImageNet ECA-Net (ResNet-101) Top 1 Accuracy 78.65% # 441
Top 5 Accuracy 94.34% # 147
Number of params 42.49M # 184
Image Classification ImageNet ECA-Net (MobileNetV2) Top 1 Accuracy 72.56% # 557
Top 5 Accuracy 90.81% # 211
Number of params 3.34M # 357

Methods