MVGamba: Unify 3D Content Generation as State Space Sequence Modeling
This addresses the issue of efficient and high-quality 3D content generation for applications in computer graphics and AI, though it is incremental as it builds on existing multi-view diffusion and Gaussian splatting methods.
The paper tackles the problem of multi-view inconsistency and blurred textures in 3D Gaussian reconstruction models by introducing MVGamba, a lightweight model based on a State Space Model, which outperforms state-of-the-art baselines in 3D content generation with only about 0.1 times the model size.
Recent 3D large reconstruction models (LRMs) can generate high-quality 3D content in sub-seconds by integrating multi-view diffusion models with scalable multi-view reconstructors. Current works further leverage 3D Gaussian Splatting as 3D representation for improved visual quality and rendering efficiency. However, we observe that existing Gaussian reconstruction models often suffer from multi-view inconsistency and blurred textures. We attribute this to the compromise of multi-view information propagation in favor of adopting powerful yet computationally intensive architectures (e.g., Transformers). To address this issue, we introduce MVGamba, a general and lightweight Gaussian reconstruction model featuring a multi-view Gaussian reconstructor based on the RNN-like State Space Model (SSM). Our Gaussian reconstructor propagates causal context containing multi-view information for cross-view self-refinement while generating a long sequence of Gaussians for fine-detail modeling with linear complexity. With off-the-shelf multi-view diffusion models integrated, MVGamba unifies 3D generation tasks from a single image, sparse images, or text prompts. Extensive experiments demonstrate that MVGamba outperforms state-of-the-art baselines in all 3D content generation scenarios with approximately only $0.1\times$ of the model size.