Deep Orientation-Aware Functional Maps: Tackling Symmetry Issues in Shape Matching

Nicolas Donati, Etienne Corman, Maks Ovsjanikov

CVPR 2022 pp. 742-751

doi:10.1109/CVPR52688.2022.00082 /cvpr/2022/donati2022cvpr-deep/

Abstract

State-of-the-art fully intrinsic network for non-rigid shape matching are unable to disambiguate between shape inner symmetries. Meanwhile, recent advances in the functional map framework allow to enforce orientation preservation using a functional representation for tangent vector field transfer, through so-called complex functional maps. Using this representation, we propose a new deep learning approach to learn orientation-aware features in a fully unsupervised setting. Our architecture is built on DiffusionNet, which makes our method robust to discretization changes, while adding a vector-field-based loss, which promotes orientation preservation without using (often unstable) extrinsic descriptors. Our source code is available at: https://github.com/nicolasdonati/DUO-FM

PDF CVPR Semantic Scholar

Cite

Text

Donati et al. "Deep Orientation-Aware Functional Maps: Tackling Symmetry Issues in Shape Matching." Conference on Computer Vision and Pattern Recognition, 2022. doi:10.1109/CVPR52688.2022.00082

Markdown

[Donati et al. "Deep Orientation-Aware Functional Maps: Tackling Symmetry Issues in Shape Matching." Conference on Computer Vision and Pattern Recognition, 2022.](https://mlanthology.org/cvpr/2022/donati2022cvpr-deep/) doi:10.1109/CVPR52688.2022.00082

BibTeX

@inproceedings{donati2022cvpr-deep,
  title     = {{Deep Orientation-Aware Functional Maps: Tackling Symmetry Issues in Shape Matching}},
  author    = {Donati, Nicolas and Corman, Etienne and Ovsjanikov, Maks},
  booktitle = {Conference on Computer Vision and Pattern Recognition},
  year      = {2022},
  pages     = {742-751},
  doi       = {10.1109/CVPR52688.2022.00082},
  url       = {https://mlanthology.org/cvpr/2022/donati2022cvpr-deep/}
}