Multi-Human Parsing 벤치마크
Multi-Human Parsing on MHP v2.0
AP 0.5
- 2017-03-20 — Mask R-CNN: AP 0.5 14.9
- 2017-05-19 — MH-Parser: AP 0.5 17.99
- 2018-04-10 — NAN: AP 0.5 25.14
- 2023-04-22 — SMP: AP 0.5 47.1
- 2023-10-13 — UniParser: AP 0.5 51.2
| Rank | Model | AP 0.5 | Paper | Code | Year |
|---|---|---|---|---|---|
| 1 | UniParser | 51.2 | UniParser: Multi-Human Parsing with Unified Correlation Representation Learning | cjm-sfw/Uniparser | 2023 |
| 2 | SMP | 47.1 | Single-stage Multi-human Parsing via Point Sets and Center-based Offsets | cjm-sfw/SMP | 2023 |
| 3 | NAN | 25.14% | Understanding Humans in Crowded Scenes: Deep Nested Adversarial Learning and A New Benchmark for Multi-Human Parsing | open-mmlab/mmpose · ZhaoJ9014/Multi-Human-Parsing | 2018 |
| 4 | MH-Parser | 17.99% | Multiple-Human Parsing in the Wild | ZhaoJ9014/Multi-Human-Parsing_MHP · ZhaoJ9014/Multi-Human-Parsing | 2017 |
| 5 | Mask R-CNN | 14.9 | Mask R-CNN | tensorflow/models · facebookresearch/detectron2 · facebookresearch/detectron · +176 | 2017 |