2D Human Pose Estimation On Human Art
Metriken
AP
AP (gt bbox)
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Vergleichstabelle
| Modellname | AP | AP (gt bbox) |
|---|---|---|
| deep-high-resolution-representation-learning | 0.417 | 0.769 |
| rtmpose-real-time-multi-person-pose | 0.311 | - |
| deep-high-resolution-representation-learning | 0.399 | 0.754 |
| explicit-box-detection-unifies-end-to-end | 0.723 | / |
| vitpose-simple-vision-transformer-baselines | 0.468 | 0.800 |
| vitpose-simple-vision-transformer-baselines | 0.381 | 0.738 |
| vitpose-simple-vision-transformer-baselines | 0.459 | 0.789 |
| unipose-detecting-any-keypoints | 0.759 | - |
| vitpose-simple-vision-transformer-baselines | 0.410 | 0.759 |
| rtmpose-real-time-multi-person-pose | - | 0.753 |