You Only Look at One Sequence: Rethinking Transformer in Vision through Object Detection

TL;DR

This paper introduces YOLOS, a simple Vision Transformer-based object detection model that achieves competitive results on COCO by treating detection as a sequence-to-sequence task without complex modifications.

Contribution

The paper demonstrates that a minimal modification Vision Transformer can perform effective object detection, challenging the need for specialized spatial inductive biases.

Findings

YOLOS-Base achieves 42.0 box AP on COCO val.

Pre-trained on ImageNet-1k, YOLOS performs competitively.

Discusses limitations of current pre-training and scaling strategies.

Abstract

Can Transformer perform 2D object- and region-level recognition from a pure sequence-to-sequence perspective with minimal knowledge about the 2D spatial structure? To answer this question, we present You Only Look at One Sequence (YOLOS), a series of object detection models based on the vanilla Vision Transformer with the fewest possible modifications, region priors, as well as inductive biases of the target task. We find that YOLOS pre-trained on the mid-sized ImageNet-1k dataset only can already achieve quite competitive performance on the challenging COCO object detection benchmark, e.g., YOLOS-Base directly adopted from BERT-Base architecture can obtain 42.0 box AP on COCO val. We also discuss the impacts as well as limitations of current pre-train schemes and model scaling strategies for Transformer in vision through YOLOS. Code and pre-trained models are available at https://github.com/hustvl/YOLOS.