Focal Attention for Long-Range Interactions in Vision Transformers

NeurIPS 2021 |

Preprint

Recently, Vision Transformer and its variants have shown great promise on various computer vision tasks. The ability to capture local and global visual dependencies through self-attention is the key to its success. But it also brings challenges due to quadratic computational overhead, especially for the high-resolution vision tasks(e.g., object detection). Many recent works have attempted to reduce the cost and improve model performance by applying either coarse-grained global attention or fine-grained local attention. However, both approaches cripple the modeling power of the original self-attention mechanism of multi-layer Transformers, leading to sub-optimal solutions. In this paper, we present focal attention, a new attention mechanism that incorporates both fine-grained local and coarse-grained global interactions. In this new mechanism, each token attends its closest surrounding tokens at the fine granularity and the tokens far away at a coarse granularity and thus can capture both short- and long-range visual dependencies efficiently and effectively. With focal attention, we propose a new variant of Vision Transformer models, called Focal Transformers, which achieve superior performance over the state-of-the-art (SoTA) Vision Transformers on a range of public image classification and object detection benchmarks. In particular, our Focal Transformer models with a moderate size of 51.1M and a large size of 89.8M achieve 83.6% and 84.0%Top-1 accuracy, respectively, on ImageNet classification at 224×224. When employed as the backbones, Focal Transformers achieve consistent and substantial improvements over the current SoTA Swin Transformers [44] across 6 different object detection methods. Our largest Focal Transformer yields58.7/59.0boxmAPs and50.9/51.3mask mAPs on COCO mini-val/test-dev, and55.4mIoU onADE20K for semantic segmentation, creating new SoTA on three of the most challenging computer vision tasks.

Publication Downloads

Focal Transformer

August 11, 2021

This is a codebase for our recently released paper "Focal Self-attention for Local-Global Interactions in Vision Transformers". It developed a new sparse self-attention mechanism called focal self-attention towards more effective and efficient vision transformers. The goal is the release the code to the public so that others can reproduce the results reported in our paper.