Object Detection On Peopleart
Metriken
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Modellname | mAP | Paper Title | Repository | ||
|---|---|---|---|---|---|
| TOOD (Task-aligned One-stage Object Detection; trained on PeopleArt and PoPArt) | 47.8 | 78.0 | 49.9 | Poses of People in Art: A Data Set for Human Pose Estimation in Digital Art History | - |
| Fast R-CNN | - | 59.0 | - | Detecting People in Artwork with CNNs | |
| FasterRCNN (trained on StyleCOCO) | 36 | 68 | 33 | Improving Object Detection in Art Images Using Only Style Transfer | |
| TOOD (Task-aligned One-stage Object Detection) | 46.1 | 75.0 | 49.0 | Poses of People in Art: A Data Set for Human Pose Estimation in Digital Art History | - |
| PVT (Pyramid Vision Transformer) | 46.5 | 76.0 | 48.4 | Poses of People in Art: A Data Set for Human Pose Estimation in Digital Art History | - |
| PVT (Pyramid Vision Transformer; trained on PeopleArt and PopArt) | 49.7 | 80.5 | 51.8 | Poses of People in Art: A Data Set for Human Pose Estimation in Digital Art History | - |
0 of 6 row(s) selected.