README.md

MIP: Modality- and Instance-aware Visual Prompts Network

ICMR 2024 Oral | Enhancing Visible-Infrared Person Re-identification with Modality- and Instance-aware Visual Prompt Learning

Project home

MIP: Modality- and Instance-aware Visual Prompt Learning

Paper

[ACM Digital Library] [arXiv]

Install

This code is based on the TransReID project. Please Refer to README_TransReID.md.

Usage

Train

bash train.sh

Evaluation

bash eval_sysu.sh
bash eval_regdb.sh

Checkpoint

You can download our models trained on SYSU-MM01 and RegDB.
BaiduNetdisk: [Checkpoints] (code: m312)
GoogleDrive: [Checkpoints]

Acknowledgement

So much thanks for codebase from TransReID.

Reference

If you find this code useful for your research, please cite our paper.

@inproceedings{10.1145/3652583.3658109,
author = {Wu, Ruiqi and Jiao, Bingliang and Wang, Wenxuan and Liu, Meng and Wang, Peng},
title = {Enhancing Visible-Infrared Person Re-identification with Modality- and Instance-aware Visual Prompt Learning},
year = {2024},
isbn = {9798400706196},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3652583.3658109},
doi = {10.1145/3652583.3658109},
booktitle = {Proceedings of the 2024 International Conference on Multimedia Retrieval},
pages = {579–588},
numpages = {10},
keywords = {cross-modality person re-identification, visible-infrared person re-identification, visual prompt learning},
location = {Phuket, Thailand},
series = {ICMR '24}
}