Learning Implicit Functions for Topology-Varying Dense 3D Shape Correspondence

Abstract

The goal of this paper is to learn dense 3D shape correspondence for topology-varying objects in an unsupervised manner. Conventional implicit functions estimate the occupancy of a 3D point given a shape latent code. Instead, our novel implicit function produces a part embedding vector for each 3D point, which is assumed to be similar to its densely corresponded point in another 3D shape of the same object category. Furthermore, we implement dense correspondence through an inverse function mapping from the part embedding to a corresponded 3D point. Both functions are jointly learned with several effective loss functions to realize our assumption, together with the encoder generating the shape latent code. During inference, if a user selects an arbitrary point on the source shape, our algorithm can automatically generate a confidence score indicating whether there is a correspondence on the target shape, as well as the corresponding semantic point if there is one. Such a mechanism inherently benefits man-made objects with different part constitutions. The effectiveness of our approach is demonstrated through unsupervised 3D semantic correspondence and shape segmentation.

arXiv commentAccepted by NeurIPS 2020

References in corpus (8)

• Learning Representations and Generative Models for 3D Point Clouds
• PIFu: Pixel-Aligned Implicit Function for High-Resolution Clothed Human Digitization
• Learning to Infer Implicit Surfaces without 3D Supervision
• GRASS: Generative Recursive Autoencoders for Shape Structures
• KeypointNet: A Large-scale 3D Keypoint Dataset Aggregated from Numerous Human Annotations
• 3D-CODED : 3D Correspondences by Deep Deformation
• Unsupervised Learning of Intrinsic Structural Representation Points
• Learning Local Shape Descriptors from Part Correspondences With Multi-view Convolutional Networks