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

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 . * Yuxin Fang and Bencheng Liao contributed equally. † Xinggang Wang is the corresponding author. This work was done when Yuxin Fang was interning at Horizon Robotics mentored by Rui Wu. 1 There are various sophisticated or hybrid architectures termed as "Vision Transformer". For disambiguation, in this paper, "Vision Transformer" and "ViT" refer to the canonical or vanilla Vision Transformer architecture proposed by Dosovitskiy et al. [21] unless specified. 35th Conference on Neural Information Processing Systems (NeurIPS 2021).