FACE: A Face-based Autoregressive Representation for High-Fidelity and Efficient Mesh Generation
This work addresses the bottleneck of inefficient mesh generation for 3D content creation, offering a scalable solution that could lower barriers in fields like computer graphics and virtual reality.
The paper tackles the problem of high computational costs in autoregressive 3D mesh generation by introducing FACE, a face-based autoregressive representation that reduces sequence length by a factor of nine and achieves a compression ratio of 0.11, halving previous state-of-the-art while maintaining high reconstruction quality.
Autoregressive models for 3D mesh generation suffer from a fundamental limitation: they flatten meshes into long vertex-coordinate sequences. This results in prohibitive computational costs, hindering the efficient synthesis of high-fidelity geometry. We argue this bottleneck stems from operating at the wrong semantic level. We introduce FACE, a novel Autoregressive Autoencoder (ARAE) framework that reconceptualizes the task by generating meshes at the face level. Our one-face-one-token strategy treats each triangle face, the fundamental building block of a mesh, as a single, unified token. This simple yet powerful design reduces the sequence length by a factor of nine, leading to an unprecedented compression ratio of 0.11, halving the previous state-of-the-art. This dramatic efficiency gain does not compromise quality; by pairing our face-level decoder with a powerful VecSet encoder, FACE achieves state-of-the-art reconstruction quality on standard benchmarks. The versatility of the learned latent space is further demonstrated by training a latent diffusion model that achieves high-fidelity, single-image-to-mesh generation. FACE provides a simple, scalable, and powerful paradigm that lowers the barrier to high-quality structured 3D content creation.