ResT: An Efficient Transformer for Visual Recognition

TL;DR

ResT introduces an efficient multi-scale vision Transformer with novel memory-efficient attention, flexible position encoding, and overlapping convolution patch embedding, achieving superior performance on image recognition tasks.

Contribution

It proposes a new Transformer backbone with memory-efficient attention, flexible position encoding, and overlapping convolution patch embedding, improving efficiency and accuracy.

Findings

Outperforms state-of-the-art backbones in image classification

Demonstrates strong results on downstream vision tasks

Shows efficiency gains over existing Transformer models

Abstract

This paper presents an efficient multi-scale vision Transformer, called ResT, that capably served as a general-purpose backbone for image recognition. Unlike existing Transformer methods, which employ standard Transformer blocks to tackle raw images with a fixed resolution, our ResT have several advantages: (1) A memory-efficient multi-head self-attention is built, which compresses the memory by a simple depth-wise convolution, and projects the interaction across the attention-heads dimension while keeping the diversity ability of multi-heads; (2) Position encoding is constructed as spatial attention, which is more flexible and can tackle with input images of arbitrary size without interpolation or fine-tune; (3) Instead of the straightforward tokenization at the beginning of each stage, we design the patch embedding as a stack of overlapping convolution operation with stride on the 2D-reshaped token map. We comprehensively validate ResT on image classification and downstream tasks. Experimental results show that the proposed ResT can outperform the recently state-of-the-art backbones by a large margin, demonstrating the potential of ResT as strong backbones. The code and models will be made publicly available at https://github.com/wofmanaf/ResT.