In this project, we explore several approaches to optimise target re-identification (re-ID) as a re-ranking problem. Our work consisted in trying different methods to re-rank the re-ID results.
We will be testing our work on Market-1501 and VeriWild
- dgl
- Pytorch
- scikit-learn
The code has been included in /extension
. To compile it:
cd extension
sh make.sh
To run reranking evaluation:
- Place dataset files under 'dataset/' folder: The dataset structure should be like:
datasets/
Market/
camids.pkl
feat.pkl
ids.pkl
- python run.py Ps: Reranking runs only on GPU
Chady Raach Mehdi Zemni Marah Gamdou