AP: 68.9AP50: 89.4AP75: 75.6Test: 76.6
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@0.2: 95.03%
PCKh-0.5: 94.3
AP: 76.7
AP: 75.5
PCK@0.2: 97.0%
PCKh-0.5: 93.9
PCKh-0.5: 92.3
PCKh-0.5: 92.3
AP: 65.2AP50: 85.9AP75: 69.5APM: 66.2
A3DP: 20.21
PCKh-0.5: 92.5
PCK@0.2: 80.5Acceleration Error: 5.4
PCKh-0.5: 93.0
PCKh-0.5: 92.0
AP: 70.0APM: 71.1AP Hard: 59.4
PCKh-0.5: 92.1
AP: 70.5AP50: 89.1AP75: 76.1AP Hard: 63.8
PCKh-0.5: 92.6
PCKh-0.5: 94.1
PCKh-0.5: 93.5
PCKh-0.5: 93.2
PCKh-0.5: 92.7