Multiscale Mesh Deformation Component Analysis with Attention-based Autoencoders

TL;DR

This paper introduces a novel attention-based autoencoder approach for automatically extracting multiscale deformation components in 3D shapes, enabling more effective shape editing and understanding.

Contribution

It presents a new stacked attention autoencoder that learns to represent deformation components at multiple scales, improving over existing local-scale methods.

Findings

Outperforms state-of-the-art methods in quantitative evaluations.

Enables coarse-to-fine shape editing.

Effectively captures multi-scale deformation components.

Abstract

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.