Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
Aber Hu bbc77f427b | 2 years ago | |
---|---|---|
datasets | 2 years ago | |
losses | 2 years ago | |
models | 2 years ago | |
utils | 2 years ago | |
.gitignore | 2 years ago | |
LICENSE | 2 years ago | |
README.md | 2 years ago | |
test.py | 2 years ago | |
test_regdb.sh | 2 years ago | |
test_sysu.sh | 2 years ago | |
train_base.py | 2 years ago | |
train_eval.py | 2 years ago | |
train_regdb.sh | 2 years ago | |
train_sysu.sh | 2 years ago |
Official Pytorch code of paper CM-NAS: Cross-Modality Neural Architecture Search for Visible-Infrared Person Re-Identification in ICCV2021.
Visible-Infrared person re-identification (VI-ReID) aims to match cross-modality pedestrian images, breaking through the limitation of single-modality person ReID in dark environment. In order to mitigate the impact of large modality discrepancy, existing works manually design various two-stream architectures to separately learn modalityspecific and modality-sharable representations. Such a manual design routine, however, highly depends on massive experiments and empirical practice, which is time consuming and labor intensive. In this paper, we systematically study the manually designed architectures, and identify that appropriately separating Batch Normalization (BN) layers is the key to bring a great boost towards crossmodality matching. Based on this observation, the essential objective is to find the optimal separation scheme for each BN layer. To this end, we propose a novel method, named Cross-Modality Neural Architecture Search (CM-NAS). It consists of a BN-oriented search space in which the standard optimization can be fulfilled subject to the cross-modality task. Equipped with the searched architecture, our method outperforms state-of-the-art counterparts in both two benchmarks, improving the Rank-1/mAP by 6.70%/6.13% on SYSU-MM01 and by 12.17%/11.23% on RegDB.
Our experiments are conducted under the following environments:
The searched configurations and the trained models can be downloaded in this link.
Dataset | Protocol | Rank-1 | mAP | Protocol | Rank-1 | mAP | Trained Model |
---|---|---|---|---|---|---|---|
SYSU-MM01 | All-Single | 61.99% | 60.02% | Indoor-Single | 67.01% | 72.95% | Google Drive |
RegDB | Vis-to-Inf | 84.54% | 80.32% | Inf-to-Vis | 82.57% | 78.31% | Google Drive |
Noet, the results may have some fluctuations caused by random spliting the datasets.
Codes will be released soon.
Before training, please download the searched configurations.
data_root
path in train_sysu.sh and run train_sysu.sh.data_root
path in train_regdb.sh and run train_regdb.sh.Before testing, please download the searched configurations and the trained models.
data_root
path in test_sysu.sh and run test_sysu.sh.data_root
path in test_regdb.sh and run test_regdb.sh.CM-NAS is released under the Apache License 2.0. Please see the LICENSE file for more information.
Please consider citing our paper in your publications if the project helps your research. BibTeX reference is as follows.
@inproceedings{Fu2021CMNAS,
title = {CM-NAS: Cross-Modality Neural Architecture Search for Visible-Infrared Person Re-Identification},
author = {Chaoyou Fu, Yibo Hu, Xiang Wu, Hailin Shi, Tao Mei and Ran He},
booktitle = {ICCV},
year = {2021}
}
This repo is based on the following repo, thank the authors a lot.
CM-NAS算法旨在对跨模态识别网络中逐层的批归一化(Batch Norm)进行自动化地搜索,选择出最优的拆分和共享批归一化组合,分别统计模态拆分和模态共享的统计量,减少模态差异,进而提升跨模态识别性能。
Python Shell
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》