AP: 70.4AP50: -AP75: -AR: -
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
DUC1-Acc@0.25m,10°: 57.1DUC1-Acc@0.5m,10°: 78.8DUC1-Acc@1.0m,10°: 88.4DUC2-Acc@0.25m,10°: 70.2
AP: 33.5AP50: 78.0AP75: 23.6AR: 37.9
AP: 34.4AP50: 78.3AP75: 24.8AR: 38.7
AP: 37.1
AP: 79.2AP50: 94.4AP75: 87.1APL: 76.1
AP: 79.5AP50: 93.6AP75: 85.9APL: 84.3
AP: 81.1AP50: 95.0AP75: 88.2APL: 86.0
AP: 68.0
AP: 55.8
DUC1-Acc@0.25m,10°: 51.5DUC1-Acc@0.5m,10°: 75.3DUC1-Acc@1.0m,10°: 86.9DUC2-Acc@0.25m,10°: 63.4
DUC1-Acc@0.25m,10°: 47.5DUC1-Acc@0.5m,10°: 72.2DUC1-Acc@1.0m,10°: 84.8DUC2-Acc@0.25m,10°: 54.2
AP: 47.87
AP: 61.6
AP: 79.5AP50: 91.9AP75: 85.8APL: 86.5
AP: 75.9
DUC1-Acc@0.25m,10°: 49.0DUC1-Acc@0.5m,10°: 68.7DUC1-Acc@1.0m,10°: 80.8DUC2-Acc@0.25m,10°: 53.4
AP: 78.9AP50: 93.8AP75: 86APL: 84.5
AP: 33.6