Multiscale Mesh Deformation Component Analysis with Attention-based Autoencoders
This work provides a more effective method for analyzing and editing 3D shapes for geometry processing and shape understanding researchers, particularly for objects with multi-scale deformations.
This paper addresses the problem of extracting multi-scale deformation components from 3D meshes. The authors propose a stacked attention-based autoencoder that learns to identify and represent these components at different scales, outperforming existing state-of-the-art methods in quantitative and qualitative evaluations.
Deformation component analysis is a fundamental problem in geometry processing and shape understanding. Existing approaches mainly extract deformation components in local regions at a similar scale while deformations of real-world objects are usually distributed in a multi-scale manner. In this paper, we propose a novel method to exact multiscale deformation components automatically with a stacked attention-based autoencoder. The attention mechanism is designed to learn to softly weight multi-scale deformation components in active deformation regions, and the stacked attention-based autoencoder is learned to represent the deformation components at different scales. Quantitative and qualitative evaluations show that our method outperforms state-of-the-art methods. Furthermore, with the multiscale deformation components extracted by our method, the user can edit shapes in a coarse-to-fine fashion which facilitates effective modeling of new shapes.