MPII

MPII Human Pose

ImagesSimplified BSDIntroduced 2014-01-01

The MPII Human Pose Dataset for single person pose estimation is composed of about 25K images of which 15K are training samples, 3K are validation samples and 7K are testing samples (which labels are withheld by the authors). The images are taken from YouTube videos covering 410 different human activities and the poses are manually annotated with up to 16 body joints.

Source: 2D/3D Pose Estimation and Action Recognition using Multitask Deep Learning Image Source: http://human-pose.mpi-inf.mpg.de/

Benchmarks

1 Image, 2*2 Stitchi/PCKh@0.2 3D/PCKh@0.2 Pose Estimation/PCKh@0.2

Related Benchmarks

MPII Cooking 2 Dataset/Action Localization/Accuracy MPII Cooking 2 Dataset/Action Localization/mIoU MPII Cooking 2 Dataset/Action Segmentation/Accuracy MPII Cooking 2 Dataset/Action Segmentation/mIoU MPII Gaze/Gaze Estimation/Angular Error MPII Human Pose/1 Image, 2*2 Stitchi/PCKh-0.5 MPII Human Pose/3D/PCKh-0.5 MPII Human Pose/Pose Estimation/PCKh-0.5 MPII Movie Description/Partial Label Learning/Accuracy MPII Movie Description/Partial Label Learning/F1-Score MPII Multi-Person/1 Image, 2*2 Stitchi/AP MPII Multi-Person/1 Image, 2*2 Stitchi/mAP@0.5 MPII Multi-Person/3D/AP MPII Multi-Person/3D/mAP@0.5 MPII Multi-Person/Multi-Person Pose Estimation/AP MPII Multi-Person/Pose Estimation/AP MPII Multi-Person/Pose Estimation/mAP@0.5 MPII Single Person/1 Image, 2*2 Stitchi/PCKh@0.1 MPII Single Person/1 Image, 2*2 Stitchi/PCKh@0.5 MPII Single Person/3D/PCKh@0.1 MPII Single Person/3D/PCKh@0.5 MPII Single Person/Pose Estimation/PCKh@0.1 MPII Single Person/Pose Estimation/PCKh@0.5 MPIIGaze/Gaze Estimation/Mean Angle Error