CrossHash

A CrossFormer-Based Hashing method

A brief intro to this work

Google slides

Dataset

We using COCO2014 NUSWIDE-TC21 MIRFlikcr25k in our experiment:

COCO2014

链接: https://pan.baidu.com/s/1D88pWPcmGRgVBra2lB3MwQ?pwd=1023 提取码: 1023

NUSWIDE-TC21

链接: https://pan.baidu.com/s/1OZk-A8sohjl69oSG0reJWg?pwd=1024 提取码: 1024

MIRFlickr25K

链接: https://pan.baidu.com/s/1WnUxKbZ4cIwxIkFOgyPjuQ 提取码: 1025

Reference for how to divide the data set ：

https://github.com/swuxyj/DeepHash-pytorch

Environment Problem (Important)

This repo uses apex and timm packages

To avoid environmental conflict we provide a well-installed conda environment you just need to follow the step shown below

Using this Python interpreter to avoid environmental conflict

链接：https://pan.baidu.com/s/1rMGOdoAi8kZxWZAWsFgh9A?pwd=1026 提取码：1026 --来自百度网盘超级会员V4的分享

Citation

@article{guo2021cmt,
  title={Cmt: Convolutional neural networks meet vision transformers},
  author={Guo, Jianyuan and Han, Kai and Wu, Han and Xu, Chang and Tang, Yehui and Xu, Chunjing and Wang, Yunhe},
  journal={arXiv preprint arXiv:2107.06263},
  year={2021}
}
@article{RelaHash,
    author={Minh, Pham Vu Thai and Viet, Nguyen Dong Duc and Son, Ngo Tung and Anh, Bui Ngoc and Jaafar, Jafreezal},
    journal={IEEE Access}, 
    title={RelaHash: Deep Hashing With Relative Position}, 
    year={2023},
    volume={11},
    number={},
    pages={30094-30108},
    doi={10.1109/ACCESS.2023.3259104}
}
@inproceedings{wang2021crossformer,
  title = {CrossFormer: A Versatile Vision Transformer Hinging on Cross-scale Attention},
  author = {Wang, Wenxiao and Yao, Lu and Chen, Long and Lin, Binbin and Cai, Deng and He, Xiaofei and Liu, Wei},
  booktitle = {International Conference on Learning Representations, {ICLR}},
  url = {https://openreview.net/forum?id=_PHymLIxuI},
  year = {2022}
}