Neural Template: Topology-aware Reconstruction and Disentangled Generation of 3D Meshes

TL;DR

This paper presents DTNet, a topology-aware neural framework for 3D mesh reconstruction and generation that disentangles topology and shape, enabling diverse and controllable 3D shape synthesis.

Contribution

It introduces a novel topology-aware neural template learning approach that decouples topology and shape for improved 3D mesh reconstruction and generation.

Findings

Produces high-quality meshes with diverse topologies.

Enables disentangled control over topology and shape.

Outperforms state-of-the-art methods in quality and diversity.

Abstract

This paper introduces a novel framework called DTNet for 3D mesh reconstruction and generation via Disentangled Topology. Beyond previous works, we learn a topology-aware neural template specific to each input then deform the template to reconstruct a detailed mesh while preserving the learned topology. One key insight is to decouple the complex mesh reconstruction into two sub-tasks: topology formulation and shape deformation. Thanks to the decoupling, DT-Net implicitly learns a disentangled representation for the topology and shape in the latent space. Hence, it can enable novel disentangled controls for supporting various shape generation applications, e.g., remix the topologies of 3D objects, that are not achievable by previous reconstruction works. Extensive experimental results demonstrate that our method is able to produce high-quality meshes, particularly with diverse topologies, as compared with the state-of-the-art methods.