3D-CODED: 3D Correspondences by Deep Deformation

By Thibault Groueix, Matthew Fisher, Vladimir G. Kim, Bryan C. Russell, Mathieu Aubry. European Conference on Computer Vision, 2018

369 citations recorded by the source corpus.

We present a new deep learning approach for matching deformable shapes by introducing Shape Deformation Networks which jointly encode 3D shapes and correspondences. This is achieved by factoring the surface representation into (i) a template, that parameterizes the surface, and (ii) a learnt global feature vector that parameterizes the transformation of the template into the input surface. By predicting this feature for a new shape, we implicitly predict correspondences between this shape and the template. We show that these correspondences can be improved by an additional step which improves the shape feature by minimizing the Chamfer distance between the input and transformed template. We demonstrate that our simple approach improves on state-of-the-art results on the difficult FAUST-inter challenge, with an average correspondence error of 2.88 cm. We show, on the TOSCA dataset, that our method is robust to many types of perturbations, and generalizes to non-human shapes. This robustness allows it to perform well on real unclean, meshes from the SCAPE dataset.

Subfields: Computational Mechanics, Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/f925f3124f48680b34087453199d15022362a26e

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

In Computer Graphics and Computer-Aided Design

In Computer Vision and Pattern Recognition

Citations in the corpus

Occupancy Flow: 4D Reconstruction by Learning Particle Dynamics (2019, methodology): “In the shape matching and interpolation experiment, we compare to nearest neighbor matching, Coherent Point Drift (CPD) [43], and 3D-Coded [24], a state-of-the-art method for finding correspondences between human shapes.”
End-to-End Human Pose and Mesh Reconstruction with Transformers (2020, methodology): “In the same spirit as positional encoding [48, 20, 11], we use a template human mesh to preserve the positional information of each query in the input sequence.”
Self-supervised Learning of Dense Shape Correspondence (2018, methodology): “A recent method based on extrinsic deformation of a null-shape was introduced in [20].”
Learning Joint Reconstruction of Hands and Manipulated Objects (2019, positive): “Following [10, 18, 67] we employ two losses that penalize irregular meshes.”