# dddm

**Repository Path**: handsome1ang/dddm

## Basic Information

- **Project Name**: dddm
- **Description**: data-driven distance metric (DDDM) for Person Re-identification
- **Primary Language**: Matlab
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2019-04-20
- **Last Updated**: 2021-11-03

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

## data-driven distance metric (DDDM) for Person Re-identification

Demo code for [Zero-Shot Person Re-identification via Cross-View Consistency](https://ieeexplore.ieee.org/document/7346474/) in IEEE Transactions on Multimedia.

The core code is in *calcMCMC2.m*. 
You can change *K*, *alpha*, *beta* for different datasets.
Run *adaptation.m* directly to evaluate the method, and you can choose different datasets and evaluate methods.


### Citation
Please cite this paper in your publications if it helps your research:
```
@article{wang2016zero,
  title={Zero-shot person re-identification via cross-view consistency},
  author={Wang, Zheng and Hu, Ruimin and Liang, Chao and Yu, Yi and Jiang, Junjun and Ye, Mang and Chen, Jun and Leng, Qingming},
  journal={IEEE Transactions on Multimedia},
  volume={18},
  number={2},
  pages={260--272},
  year={2016},
  publisher={IEEE}
}
```

Contact: wangz@nii.ac.jp