BLSM: A Bone-Level Skinned Model of the Human Mesh

Abstract

We introduce BLSM, a bone-level skinned model of the human body mesh where bone scales are set prior to template synthesis, rather than the common, inverse practice. BLSM first sets bone lengths and joint angles to specify the skeleton, then specifies identity-specific surface variation, and finally bundles them together through linear blend skinning. We design these steps by constraining the joint angles to respect the kinematic constraints of the human body and by using accurate mesh convolution-based networks to capture identity-specific surface variation. We provide quantitative results on the problem of reconstructing a collection of 3D human scans, and show that we obtain improvements in reconstruction accuracy when comparing to a SMPL-type baseline. Our decoupled bone and shape representation also allows for out-of-box integration with standard graphics packages like Unity, facilitating full-body AR effects and image-driven character animation. Additional results and demos are available from the project webpage: http://arielai.com/blsm

Cite

Text

Wang et al. "BLSM: A Bone-Level Skinned Model of the Human Mesh." Proceedings of the European Conference on Computer Vision (ECCV), 2020. doi:10.1007/978-3-030-58558-7_1

Markdown

[Wang et al. "BLSM: A Bone-Level Skinned Model of the Human Mesh." Proceedings of the European Conference on Computer Vision (ECCV), 2020.](https://mlanthology.org/eccv/2020/wang2020eccv-blsm/) doi:10.1007/978-3-030-58558-7_1

BibTeX

@inproceedings{wang2020eccv-blsm,
  title     = {{BLSM: A Bone-Level Skinned Model of the Human Mesh}},
  author    = {Wang, Haoyang and Güler, Riza Alp and Kokkinos, Iasonas and Papandreou, George and Zafeiriou, Stefanos},
  booktitle = {Proceedings of the European Conference on Computer Vision (ECCV)},
  year      = {2020},
  doi       = {10.1007/978-3-030-58558-7_1},
  url       = {https://mlanthology.org/eccv/2020/wang2020eccv-blsm/}
}