TetSphere Splatting: Representing High-Quality Geometry with Lagrangian Volumetric Meshes

TL;DR

TetSphere Splatting introduces a novel Lagrangian volumetric mesh-based method for high-quality 3D shape modeling, effectively addressing common mesh issues and demonstrating versatility in reconstruction and generative tasks.

Contribution

It presents TetSphere Splatting, a new approach using tetrahedral meshes for 3D shape representation, improving mesh quality and integration into generative modeling.

Findings

Superior mesh quality in multi-view and single-view reconstruction

Maintains competitive accuracy with state-of-the-art methods

Effective in image-to-3D and text-to-3D generative tasks

Abstract

We introduce TetSphere Splatting, a Lagrangian geometry representation designed for high-quality 3D shape modeling. TetSphere splatting leverages an underused yet powerful geometric primitive -- volumetric tetrahedral meshes. It represents 3D shapes by deforming a collection of tetrahedral spheres, with geometric regularizations and constraints that effectively resolve common mesh issues such as irregular triangles, non-manifoldness, and floating artifacts. Experimental results on multi-view and single-view reconstruction highlight TetSphere splatting's superior mesh quality while maintaining competitive reconstruction accuracy compared to state-of-the-art methods. Additionally, TetSphere splatting demonstrates versatility by seamlessly integrating into generative modeling tasks, such as image-to-3D and text-to-3D generation.