Exploring Self-attention for Image Recognition

Abstract

Recent work has shown that self-attention can serve as a basic building block for image recognition models. We explore variations of self-attention and assess their effectiveness for image recognition. We consider two forms of self-attention. One is pairwise self-attention, which generalizes standard dot-product attention and is fundamentally a set operator. The other is patchwise self-attention, which is strictly more powerful than convolution. Our pairwise self-attention networks match or outperform their convolutional counterparts, and the patchwise models substantially outperform the convolutional baselines. We also conduct experiments that probe the robustness of learned representations and conclude that self-attention networks may have significant benefits in terms of robustness and generalization.

CVPR 2020, Code available at https://github.com/hszhao/SAN

References in corpus (5)

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
• Large Scale GAN Training for High Fidelity Natural Image Synthesis
• Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour
• Pay Less Attention with Lightweight and Dynamic Convolutions

Cited by in corpus (5)

• Visual Transformers: Token-based Image Representation and Processing for Computer Vision
• On the Integration of Self-Attention and Convolution
• Multi-Scale Feature Aggregation by Cross-Scale Pixel-to-Region Relation Operation for Semantic Segmentation
• Auto-decoding Graphs
• Graph Attention Networks with Positional Embeddings