Home

Awesome

ACmix

This repo contains the official PyTorch code and pre-trained models for ACmix.

Update

Introduction

main

We explore a closer relationship between convolution and self-attention in the sense of sharing the same computation overhead (1×1 convolutions), and combining with the remaining lightweight aggregation operations.

Results

image-20211208195403247

Pretrained Models

Backbone ModelsParamsFLOPsTop-1 AccLinks
ResNet-2610.6M2.3G76.1 (+2.5)In process
ResNet-3814.6M2.9G77.4 (+1.4)In process
ResNet-5018.6M3.6G77.8 (+0.9)In process
SAN-1012.1M1.9G77.6 (+0.5)In process
SAN-1516.6M2.7G78.4 (+0.4)In process
SAN-1921.2M3.4G78.7 (+0.5)In process
PVT-T13M2.0G78.0 (+2.9)In process
PVT-S25M3.9G81.7 (+1.9)In process
Swin-T30M4.6G81.9 (+0.6)Tsinghua Cloud / Google Drive
Swin-S51M9.0G83.5 (+0.5)Tsinghua Cloud / Google Drive

Get Started

Please go to the folder ResNet, Swin-Transformer for specific docs.

Contact

If you have any question, please feel free to contact the authors. Xuran Pan: pxr18@mails.tsinghua.edu.cn.

Acknowledgment

Our code is based on SAN, PVT, and Swin Transformer.

Citation

If you find our work is useful in your research, please consider citing:

@misc{pan2021integration,
      title={On the Integration of Self-Attention and Convolution}, 
      author={Xuran Pan and Chunjiang Ge and Rui Lu and Shiji Song and Guanfu Chen and Zeyi Huang and Gao Huang},
      year={2021},
      eprint={2111.14556},
      archivePrefix={arXiv},
      primaryClass={cs.CV}
}