Person Re-Identification
Person Re-Identification is a computer vision task in which the goal is to match a person's identity across different cameras or locations in a video or image sequence. It involves detecting and tracking a person and then using features such as appearance, body shape, and clothing to match their identity in different frames. The goal is to associate the same person across multiple non-overlapping camera views in a robust and efficient manner.
Papers
Showing 1–10 of 1488 papers
All datasetsMarket-1501DukeMTMC-reIDMSMT17Occluded-DukeMTMCMarket-1501-CCUHK03 detectedCUHK03 labeledMARSCUHK03LTCCPRCCPRID2011
Benchmark Results
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | Weakly Pre-training (ResNet101+RK) | MAP | 89.29 | — | Unverified |
| 2 | Top-DB-Net + RK | MAP | 88.5 | — | Unverified |
| 3 | DiP (without RK) | MAP | 85.7 | — | Unverified |
| 4 | LightMBN (w/o ReRank) | MAP | 85.1 | — | Unverified |
| 5 | Deep Miner (w/o ReRank) | MAP | 84.7 | — | Unverified |
| 6 | FPB | MAP | 83.8 | — | Unverified |
| 7 | MPN (without re-ranking) | MAP | 81.1 | — | Unverified |
| 8 | PLR-OSNet | MAP | 80.5 | — | Unverified |
| 9 | Pyramid (CVPR' 19) | MAP | 76.9 | — | Unverified |
| 10 | BDB (ICCV'19) | MAP | 76.7 | — | Unverified |