Semi Supervised Human Pose Estimation On
Metriken
mAP
mAP@0.5
mAP@0.75
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Modellname | mAP | mAP@0.5 | mAP@0.75 | Paper Title | Repository |
|---|---|---|---|---|---|
| HRNet-W32 (trained on COCO 2017 with 0 % style-transferred Material) | 25.25 | 31.73 | 28.10 | Semi-supervised Human Pose Estimation in Art-historical Images | |
| HRNet-W32 (trained on COCO 2017 with 100 % style-transferred Material) | 25.18 | 31.67 | 28.13 | Semi-supervised Human Pose Estimation in Art-historical Images | |
| HRNet-W32 (trained on PeopleArt) | 29.71 | 36.37 | 32.72 | Semi-supervised Human Pose Estimation in Art-historical Images | |
| HRNet-W32 (trained on COCO 2017 with 50 % style-transferred Material) | 24.13 | 30.52 | 26.65 | Semi-supervised Human Pose Estimation in Art-historical Images | |
| HRNet-W32 | 52.58 | 63.92 | 57.35 | Semi-supervised Human Pose Estimation in Art-historical Images |
0 of 5 row(s) selected.