Liming Zhao, Xi Li, Yueting Zhuang, and Jingdong Wang. “Deeply-Learned Part-Aligned Representations for Person Re-Identification.” Proceedings of the International Conference on Computer Vision (ICCV), 2017. (paper)
@InProceedings{Zhao_2017_ICCV,
author = {Zhao, Liming and Li, Xi and Zhuang, Yueting and Wang, Jingdong},
title = {Deeply-Learned Part-Aligned Representations for Person Re-Identification},
booktitle = {The IEEE International Conference on Computer Vision (ICCV)},
pages = {3219-3228},
year = {2017}
}Contact: Liming Zhao (zlmzju@gmail.com)
Use my Caffe for using triplet loss layer.
Run the demo code demo/demo.ipynb to see an example usage.
Run train.sh in the train folder to train the model.
The datasets are placed in the dataset folder, you can download the archived data from here. Training list can be generated by using the code provided in the archieved data.
Use Caffe for implementation, please refer to the Caffe project website for installation.
The protocal file in proto folder is written in python.
The actual training scripts and protocal files will be generated in the train folder.