Unsupervised Deep Learning for Structured Shape Matching

12/10/2018
by   Jean-Michel Roufosse, et al.
0

We present a novel method for computing correspondences across shapes using unsupervised learning. Our method allows to compute a non-linear transformation of given descriptor functions, while optimizing for global structural properties of the resulting maps, such as their bijectivity or approximate isometry. To this end, we use the functional maps framework, and build upon the recently proposed FMNet architecture for descriptor learning. Unlike the method proposed in that work, however, we show that learning can be done in a purely unsupervised setting, without having access to any ground truth correspondences. This results in a very general shape matching method, which can be used to establish correspondences within shape collections or even just a single shape pair, without any prior information. We demonstrate on a wide range of challenging benchmarks, that our method leads to significant improvement compared to the existing axiomatic methods and achieves comparable, and in some cases superior results to even the supervised learning techniques.

READ FULL TEXT

page 6

page 8

page 11

research
07/20/2022

Unsupervised Deep Multi-Shape Matching

3D shape matching is a long-standing problem in computer vision and comp...
research
04/27/2023

Unsupervised Learning of Robust Spectral Shape Matching

We propose a novel learning-based approach for robust 3D shape matching....
research
10/23/2020

Unsupervised Dense Shape Correspondence using Heat Kernels

In this work, we propose an unsupervised method for learning dense corre...
research
12/19/2020

Unsupervised Scale-Invariant Multispectral Shape Matching

Alignment between non-rigid stretchable structures is one of the hardest...
research
10/05/2022

Smooth Non-Rigid Shape Matching via Effective Dirichlet Energy Optimization

We introduce pointwise map smoothness via the Dirichlet energy into the ...
research
12/27/2018

Eyes on the Prize: Improved Registration via Forward Propagation

We develop a robust method for improving pairwise correspondences for a ...
research
05/10/2022

Non-Isometric Shape Matching via Functional Maps on Landmark-Adapted Bases

We propose a principled approach for non-isometric landmark-preserving n...

Please sign up or login with your details

Forgot password? Click here to reset