Semantic-Aligned Learning with Collaborative Refinement for Unsupervised VI-ReID (IJCV 2025)

February 8, 2026 · View on GitHub

Pytorch Code of our paper ``Semantic-Aligned Learning with Collaborative Refinement for Unsupervised VI-ReID" for Cross-Modality Person Re-Identification (Visible Thermal Re-ID) on RegDB dataset [1] and SYSU-MM01 dataset [2].

Our paper is accepted by International Journal of Computer Vision (IJCV) 2025, available at:
Semantic-Aligned Learning with Collaborative Refinement for Unsupervised VI-ReID (SALCR)

We adopt the two-stream network structure introduced in [3]. ResNet50 is adopted as the backbone.

*Both of these two datasets may have some fluctuation due to random spliting. The results might be better by finetuning the hyper-parameters.

1. Prepare the datasets.

  • (1) RegDB Dataset [1]: The RegDB dataset can be downloaded from this website by submitting a copyright form.

    • (Named: "Dongguk Body-based Person Recognition Database (DBPerson-Recog-DB1)" on their website).

  • (2) SYSU-MM01 Dataset [2]: The SYSU-MM01 dataset can be downloaded from this website.

    • run python pre_process_sysu.py to pepare the dataset, the training data will be stored in ".npy" format.

2. Requirements

  • faiss_cpu==1.7.4
  • infomap==2.7.1
  • matplotlib==3.8.0
  • numpy==1.26.2
  • Pillow==10.0.1
  • Pillow==10.1.0
  • scikit_learn==1.3.1
  • scipy==1.11.4
  • seaborn==0.13.0
  • torch==2.0.1
  • torchvision==0.15.2
  • tqdm==4.66.1

3. Download.

A baseline model trained under the DCL framework[4] can be downloaded from this website in baseline_model.pth.tar.

A final model trained under our SALCR framework can be downloaded from this website in SALCR_model.pth.tar

4. Training.

Train a model on SYSU-MM01:

python3 main.py --dataset sysu --batch-size 12 --num_pos 12 --eps 0.6 --pretrained False

Train a model pretrained by the DCL [4] framework on SYSU-MM01:

python3 main.py --dataset sysu --batch-size 12 --num_pos 12 --eps 0.6 --pretrained True

Train a model on RegDB:

python3 main.py --dataset regdb --batch-size 12 --num_pos 12 --eps 0.3 --pretrained False

Train a model pretrained by the DCL [4] framework on RegDB:

python3 main.py --dataset regdb --batch-size 12 --num_pos 12 --eps 0.3 --pretrained True

You may need mannully define the data path first.

Parameters: More parameters can be found in the script.

5. Testing.

Test a model on SYSU-MM01:

python tester.py --sysu_model_dir 'model_path'  --dataset sysu

Test a model on RegDB:

python tester.py --sysu_model_dir 'model_path'  --dataset regdb
  • --dataset: which dataset "sysu" or "regdb".

[1] D. T. Nguyen, H. G. Hong, K. W. Kim, and K. R. Park. Person recognition system based on a combination of body images from visible light and thermal cameras. Sensors, 17(3):605, 2017.

[2] A. Wu, W.-s. Zheng, H.-X. Yu, S. Gong, and J. Lai. Rgb-infrared crossmodality person re-identification. In IEEE International Conference on Computer Vision (ICCV), pages 5380–5389, 2017.

[3] M. Ye, X. Lan, Z. Wang, and P. C. Yuen. Bi-directional Center-Constrained Top-Ranking for Visible Thermal Person Re-Identification. IEEE Transactions on Information Forensics and Security (TIFS), 2019.

[4] Bin Yang, Mang Ye, Jun Chen, and Zesen Wu. 2022. Augmented Dual-Contrastive Aggregation Learning for Unsupervised Visible-Infrared Person Re-Identification. In Proceedings of the 30th ACM International Conference on Multimedia (MM '22). Association for Computing Machinery, New York, NY, USA, 2843–2851. https://doi.org/10.1145/3503161.3548198

Contact: lfhe@stu.xidian.edu.cn