Skip to yearly menu bar Skip to main content


Poster

MeshLLM: Empowering Large Language Models to Progressively Understand and Generate 3D Mesh

Shuangkang Fang · I-Chao Shen · Yufeng Wang · Yi-Hsuan Tsai · Yi Yang · Shuchang Zhou · Wenrui Ding · Takeo Igarashi · Ming-Hsuan Yang


Abstract:

We present MeshLLM, a novel framework that leverages large language models (LLMs) to understand and generate text-serialized 3D meshes. Our approach addresses key limitations in existing methods, including the limited dataset scale when catering to LLMs' token length and the loss of 3D structural information during mesh serialization. We introduce a Primitive-Mesh decomposition strategy, which divides 3D meshes into structurally meaningful subunits. This enables the creation of a large-scale dataset with 1500k+ samples, almost 50x larger than previous methods, which aligns better with the LLM scaling law principles. Furthermore, we propose inferring face connectivity from vertices and local mesh assembly training strategies, significantly enhancing the LLMs' ability to capture mesh topology and spatial structures. Experiments show that MeshLLM outperforms the state-of-the-art LLaMA-Mesh in both mesh generation quality and shape understanding, highlighting its great potential in processing text-serialized 3D meshes.

Live content is unavailable. Log in and register to view live content