Ranking - 3D Reconstruction - All Frames

Method Time Description 3D Joints Translation Error 3D Joints MPJPE 3D Joints MPJPE-PA GHUM MPVPE GHUM MPVPE-PA SMPLX MPVPE SMPLX MPVPE-PA
Cliff 06/08/2023 - 15:24
title={Cliff: Carrying location information in full frames into human pose and shape estimation},
author={Li, Zhihao and Liu, Jianzhuang and Zhang, Zhensong and Xu, Songcen and Yan, Youliang},
booktitle={Computer Vision--ECCV 2022: 17th European Conference, Tel Aviv, Israel, October 23--27, 2022, Proceedings, Part V},
975.53 68.96 40.34 67.05 37.55 64.18 32.57
CRMH 06/08/2023 - 15:25
title={Coherent reconstruction of multiple humans from a single image},
author={Jiang, Wen and Kolotouros, Nikos and Pavlakos, Georgios and Zhou, Xiaowei and Daniilidis, Kostas},
booktitle={Proceedings of the IEEE/CVF conference on computer vision and pattern recognition},
3485.88 77.83 46.60 85.21 49.79 79.28 43.49
REMIPS 06/10/2023 - 16:48
This is a weakly-supervised method. No 3d ground truth data (joints, rotation angles, vertices) was used in training. The CHI3D training subset is not used in any form to train this method.
PDF: https://proceedings.neurips.cc/paper/2021/file/a1a2c3fed88e9b3ba5bc3625c074a04e-Paper.pdf
title={REMIPS: Physically consistent 3d reconstruction of multiple interacting people under weak supervision},
author={Fieraru, Mihai and Zanfir, Mihai and Szente, Teodor and Bazavan, Eduard and Olaru, Vlad and Sminchisescu, Cristian},
journal={Advances in Neural Information Processing Systems},
589.88 91.15 57.00 92.23 58.01 84.05 49.93