Learning Transferable Kinematic Dictionary for 3D Human Pose and Shape Reconstruction [article]

Ze Ma, Yifan Yao, Pan Ji, Chao Ma
2021 arXiv   pre-print
Estimating 3D human pose and shape from a single image is highly under-constrained. To address this ambiguity, we propose a novel prior, namely kinematic dictionary, which explicitly regularizes the solution space of relative 3D rotations of human joints in the kinematic tree. Integrated with a statistical human model and a deep neural network, our method achieves end-to-end 3D reconstruction without the need of using any shape annotations during the training of neural networks. The kinematic
more » ... ctionary bridges the gap between in-the-wild images and 3D datasets, and thus facilitates end-to-end training across all types of datasets. The proposed method achieves competitive results on large-scale datasets including Human3.6M, MPI-INF-3DHP, and LSP, while running in real-time given the human bounding boxes.
arXiv:2104.00953v2 fatcat:xd6tytqdlvb4hhbbddqc3s3h3a