Unsupervised Dense Deformation Embedding Network for Template-Free Shape Correspondence

TL;DR

This paper introduces UD^2E-Net, an unsupervised deep learning model that predicts dense shape deformations using local features and an autoencoder, avoiding dense annotations and capturing local details effectively.

Contribution

The paper proposes a novel unsupervised network with an autoencoder and deformation graph for accurate shape correspondence without dense supervision.

Findings

Outperforms state-of-the-art unsupervised methods by 24%.

Surpasses supervised methods by 13% on the Faust Intra challenge.

Effectively captures local geometric details of shapes.

Abstract

Shape correspondence from 3D deformation learning has attracted appealing academy interests recently. Nevertheless, current deep learning based methods require the supervision of dense annotations to learn per-point translations, which severely overparameterize the deformation process. Moreover, they fail to capture local geometric details of original shape via global feature embedding. To address these challenges, we develop a new Unsupervised Dense Deformation Embedding Network (i.e., UD^2E-Net), which learns to predict deformations between non-rigid shapes from dense local features. Since it is non-trivial to match deformation-variant local features for deformation prediction, we develop an Extrinsic-Intrinsic Autoencoder to frst encode extrinsic geometric features from source into intrinsic coordinates in a shared canonical shape, with which the decoder then synthesizes corresponding target features. Moreover, a bounded maximum mean discrepancy loss is developed to mitigate the distribution divergence between the synthesized and original features. To learn natural deformation without dense supervision, we introduce a coarse parameterized deformation graph, for which a novel trace and propagation algorithm is proposed to improve both the quality and effciency of the deformation. Our UD^2E-Net outperforms state-of-the-art unsupervised methods by 24% on Faust Inter challenge and even supervised methods by 13% on Faust Intra challenge.