Direct Learning of Mesh and Appearance via 3D Gaussian Splatting

TL;DR

This paper introduces an end-to-end learnable scene model that combines 3D Gaussian Splatting with explicit mesh geometry, improving reconstruction efficiency, rendering quality, and scene manipulation capabilities.

Contribution

It presents a novel method that jointly learns mesh and appearance in a differentiable framework, integrating 3D Gaussian Splatting with explicit mesh representation.

Findings

Enhanced reconstruction efficiency and rendering quality.

Ability to manipulate scenes via explicit mesh.

Effective end-to-end learning of geometry and appearance.

Abstract

Accurately reconstructing a 3D scene including explicit geometry information is both attractive and challenging. Geometry reconstruction can benefit from incorporating differentiable appearance models, such as Neural Radiance Fields and 3D Gaussian Splatting (3DGS). However, existing methods encounter efficiency issues due to indirect geometry learning and the paradigm of separately modeling geometry and surface appearance. In this work, we propose a learnable scene model that incorporates 3DGS with an explicit geometry representation, namely a mesh. Our model learns the mesh and appearance in an end-to-end manner, where we bind 3D Gaussians to the mesh faces and perform differentiable rendering of 3DGS to obtain photometric supervision. The model creates an effective information pathway to supervise the learning of both 3DGS and mesh. Experimental results demonstrate that the learned scene model not only improves efficiency and rendering quality but also enables manipulation via the explicit mesh. In addition, our model has a unique advantage in adapting to scene updates, thanks to the end-to-end learning of both mesh and appearance.