Neural Deformation Graphs for Globally-consistent Non-rigid Reconstruction

TL;DR

This paper presents Neural Deformation Graphs, a deep learning-based method for globally-consistent non-rigid 3D reconstruction that does not rely on object-specific structures, enabling robust tracking of fast or disconnected motions.

Contribution

The introduction of neural deformation graphs that are object-agnostic and optimized globally for non-rigid reconstruction from depth sequences, with self-supervised training.

Findings

Outperforms state-of-the-art methods in reconstruction quality.

Achieves 64% better reconstruction accuracy.

Attains 62% improvement in deformation tracking.

Abstract

We introduce Neural Deformation Graphs for globally-consistent deformation tracking and 3D reconstruction of non-rigid objects. Specifically, we implicitly model a deformation graph via a deep neural network. This neural deformation graph does not rely on any object-specific structure and, thus, can be applied to general non-rigid deformation tracking. Our method globally optimizes this neural graph on a given sequence of depth camera observations of a non-rigidly moving object. Based on explicit viewpoint consistency as well as inter-frame graph and surface consistency constraints, the underlying network is trained in a self-supervised fashion. We additionally optimize for the geometry of the object with an implicit deformable multi-MLP shape representation. Our approach does not assume sequential input data, thus enabling robust tracking of fast motions or even temporally disconnected recordings. Our experiments demonstrate that our Neural Deformation Graphs outperform state-of-the-art non-rigid reconstruction approaches both qualitatively and quantitatively, with 64% improved reconstruction and 62% improved deformation tracking performance.