Test AP: 32.8Validation AP: 40.0
Papers with Code · CC BY-SA 4.0
Pose Estimation
Pose Estimation is a computer vision task where the goal is to detect the position and orientation of a person or an object. Usually, this is done by predicting the location of specific keypoints like hands, head, elbows, etc. in case of Human Pose Estimation.
A common benchmark for this task is MPII Human Pose
<span style="color:grey; opacity: 0.6">( Image credit: Real-time 2D Multi-Person Pose Estimation on CPU: Lightweight OpenPose )</span>
논문
Test AP: 48.3Validation AP: 48.6
Test AP: 42.5Validation AP: 42.0
Test AP: 33.3Validation AP: 41.0
AP: 76.5
Test AP: 45.6
MAE yaw (º): 1.62MAE pitch (º): 1.80MAE roll (º): 1.24
Test AP: 45.0Validation AP: 46.1
Mean PCK@0.2: 91.9
Mean mAP: 86.92
Mean PCK@0.2: 99.0Mean PCK@0.1: 94.6Mean PCK@0.05: 80.6
Test AP: 36.0Validation AP: 41.8
Mean PCK@0.2: 80.5Mean PCK@0.1: 58.3
Mean PCK@0.2: 99.6Mean PCK@0.1: 98.3Mean PCK@0.05: 91.9
Mean PCK@0.2: 93.6
MAE mean (º): 1.56
0..5sec: Jaau
Test AP: 87.0Validation AP: 86.0
Test AP: 47.2Validation AP: 47.7
Test AP: 30.7Validation AP: 38.8
Test AP: 80.3Validation AP: 80.5
MAE mean (º): 1.29MAE yaw (º): 1.41MAE pitch (º): 1.70MAE roll (º): 0.77
Validation AP: 62.3
Test AP: 93.3Validation AP: 92.8