You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

1.7 KiB

HRFormer: High-Resolution Vision Transformer for Dense Predict

HRFormer (NIPS'2021)
@article{yuan2021hrformer,
  title={HRFormer: High-Resolution Vision Transformer for Dense Predict},
  author={Yuan, Yuhui and Fu, Rao and Huang, Lang and Lin, Weihong and Zhang, Chao and Chen, Xilin and Wang, Jingdong},
  journal={Advances in Neural Information Processing Systems},
  volume={34},
  year={2021}
}

Abstract

We present a High-Resolution Transformer (HRFormer) that learns high-resolution representations for dense prediction tasks, in contrast to the original Vision Transformer that produces low-resolution representations and has high memory and computational cost. We take advantage of the multi-resolution parallel design introduced in high-resolution convolutional networks (HRNet), along with local-window self-attention that performs self-attention over small non-overlapping image windows, for improving the memory and computation efficiency. In addition, we introduce a convolution into the FFN to exchange information across the disconnected image windows. We demonstrate the effectiveness of the HighResolution Transformer on both human pose estimation and semantic segmentation tasks, e.g., HRFormer outperforms Swin transformer by 1.3 AP on COCO pose estimation with 50% fewer parameters and 30% fewer FLOPs. Code is available at: https://github.com/HRNet/HRFormer