MARS: Mesh AutoRegressive Model for 3D Shape Detailization
This addresses the challenge of generating detailed 3D meshes consistently across categories for applications in computer graphics and 3D modeling, representing a strong specific gain rather than a foundational advance.
The paper tackles the problem of 3D shape detailization, where existing GAN-based methods often fail to generalize across categories and ensure shape consistency, by introducing MARS, a mesh autoregressive model that achieves state-of-the-art performance on a benchmark, enhancing realism and preserving shape integrity.
State-of-the-art methods for mesh detailization predominantly utilize Generative Adversarial Networks (GANs) to generate detailed meshes from coarse ones. These methods typically learn a specific style code for each category or similar categories without enforcing geometry supervision across different Levels of Detail (LODs). Consequently, such methods often fail to generalize across a broader range of categories and cannot ensure shape consistency throughout the detailization process. In this paper, we introduce MARS, a novel approach for 3D shape detailization. Our method capitalizes on a novel multi-LOD, multi-category mesh representation to learn shape-consistent mesh representations in latent space across different LODs. We further propose a mesh autoregressive model capable of generating such latent representations through next-LOD token prediction. This approach significantly enhances the realism of the generated shapes. Extensive experiments conducted on the challenging 3D Shape Detailization benchmark demonstrate that our proposed MARS model achieves state-of-the-art performance, surpassing existing methods in both qualitative and quantitative assessments. Notably, the model's capability to generate fine-grained details while preserving the overall shape integrity is particularly commendable.