Attentive Normalization

ECCV 2020  ·  Xilai Li, Wei Sun, Tianfu Wu ·

In state-of-the-art deep neural networks, both feature normalization and feature attention have become ubiquitous. % with significant performance improvement shown in a vast amount of tasks. They are usually studied as separate modules, however. In this paper, we propose a light-weight integration between the two schema and present Attentive Normalization (AN). Instead of learning a single affine transformation, AN learns a mixture of affine transformations and utilizes their weighted-sum as the final affine transformation applied to re-calibrate features in an instance-specific way. The weights are learned by leveraging channel-wise feature attention. In experiments, we test the proposed AN using four representative neural architectures in the ImageNet-1000 classification benchmark and the MS-COCO 2017 object detection and instance segmentation benchmark. AN obtains consistent performance improvement for different neural architectures in both benchmarks with absolute increase of top-1 accuracy in ImageNet-1000 between 0.5\% and 2.7\%, and absolute increase up to 1.8\% and 2.2\% for bounding box and mask AP in MS-COCO respectively. We observe that the proposed AN provides a strong alternative to the widely used Squeeze-and-Excitation (SE) module. The source codes are publicly available at https://github.com/iVMCL/AOGNet-v2 (the ImageNet Classification Repo) and https://github.com/iVMCL/AttentiveNorm\_Detection (the MS-COCO Detection and Segmentation Repo).

PDF Abstract ECCV 2020 PDF ECCV 2020 Abstract

Datasets


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Instance Segmentation COCO minival Mask R-CNN-FPN (AOGNet-40M) mask AP 40.2 # 71
AP50 63.2 # 10
AP75 43.3 # 12
Object Detection COCO minival Mask R-CNN-FPN (AOGNet-40M) box AP 44.9 # 109
AP50 66.2 # 34
AP75 49.1 # 32
Image Classification ImageNet AOGNet-40M-AN Top 1 Accuracy 81.87% # 552
GFLOPs 7.51 # 260

Methods