Human Pose Estimation on MSCOCO Keypoint Challenge 2016 (test)
72.3APRMPE++ (ours++)
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| RMPE++ (ours++)Human Detector=Faster R-CNN with Soft-NMS [5], Pose Estimator=PyraNet [45], Input Size=320x256, Ensembling=false2016.12 | 72.3 | 89.2 | 79.1 | 68 | 78.6 | |
| MegviiEnsembling=false2016.12 | 72.1 | 91.4 | 80 | 68.7 | 77.2 | |
| G-RMIEnsembling=false2016.12 | 68.5 | 87.1 | 75.5 | 65.8 | 73.3 | |
| Mask R-CNNEnsembling=false2016.12 | 63.1 | 87.3 | 68.7 | 57.8 | 71.4 | |
| CMU-PoseEnsembling=false2016.12 | 61.8 | 84.9 | 67.5 | 57.1 | 68.2 | |
| RMPE (ours)Ensembling=false2016.12 | 61.8 | 83.7 | 69.8 | 58.6 | 67.6 |