Unite the People: Closing the Loop Between 3D and 2D Human Representations

Benchmark Model Rank Results
3d-human-pose-estimation-on-humaneva-iSMPLify (dense)#15Mean Reconstruction Error (mm): 74.5
monocular-3d-human-pose-estimation-on-human3SMPLify (dense)#38Use Video Sequence: NoFrames Needed: 1Need Ground Truth 2D Pose: No
monocular-3d-human-pose-estimation-on-human3SMPLify (dense)#45PA-MPJPE: 80.7