DeepMesh: Auto-Regressive Artist-mesh Creation with Reinforcement Learning
This work solves the problem of efficient and detailed mesh generation for 3D modeling and rendering applications, representing an incremental improvement with novel method integrations.
The paper tackles the problem of generating high-quality triangle meshes in 3D applications by addressing limitations like limited face counts and incompleteness in auto-regressive methods, resulting in a framework that outperforms state-of-the-art methods in precision and quality.
Triangle meshes play a crucial role in 3D applications for efficient manipulation and rendering. While auto-regressive methods generate structured meshes by predicting discrete vertex tokens, they are often constrained by limited face counts and mesh incompleteness. To address these challenges, we propose DeepMesh, a framework that optimizes mesh generation through two key innovations: (1) an efficient pre-training strategy incorporating a novel tokenization algorithm, along with improvements in data curation and processing, and (2) the introduction of Reinforcement Learning (RL) into 3D mesh generation to achieve human preference alignment via Direct Preference Optimization (DPO). We design a scoring standard that combines human evaluation with 3D metrics to collect preference pairs for DPO, ensuring both visual appeal and geometric accuracy. Conditioned on point clouds and images, DeepMesh generates meshes with intricate details and precise topology, outperforming state-of-the-art methods in both precision and quality. Project page: https://zhaorw02.github.io/DeepMesh/