PCK: 73.4%
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
PCK: 90.5%
PCK: 90.1%
AP: 63.3AP50: 86.7AP75: 69.4APL: 71.2
PCK: 90.8%
PCK: 90.7%
PCK: 90.5%
PCK: 94.5%
PCK: 93.9%
PCK: 87.5%
PCK: 93.9%
AP: 63.1AP50: 87.3AP75: 68.7APL: 71.4
PCK: 75.7
PCK: 92.6%
PCK: 99.5%
AP: -AP50: -AP75: -APL: -
AP: 64.2AP50: 86.2AP75: 70.1APL: 68.8
AP: 61.8AP50: 84.9AP75: 67.5APL: 68.2
AP: 63.8AP50: 88.4AP75: 70.4APL: 71.7
AP: 78.5AP Hard: 72.3AP Easy: 83.9AP Medium: 79.0
AP: 64.4AP50: 85.7AP75: 70.7APL: 69.8
AP: 61.8AP50: 83.7AP75: 69.8APL: 67.6
PCK: 94%
MAE mean (º): 2.39MAE yaw (º): 3.23MAE pitch (º): 2.24MAE roll (º): 1.71
PCK: 94.8%
PCK: 83.6
PCK: 94.5%
AP: 78.3AP50: 85.3AP75: 81.4APM: 86.6
PCK: 79.4
AP50: 90.3