Efficient 3D Shape Generation via Diffusion Mamba with Bidirectional SSMs

TL;DR

This paper introduces Diffusion Mamba (DiM-3D), a scalable and efficient 3D shape generation model that leverages the Mamba architecture to overcome the limitations of traditional diffusion transformers, achieving high-quality results at high resolutions.

Contribution

We propose a novel diffusion architecture for 3D shape generation that replaces attention with the Mamba architecture, enabling linear complexity and improved scalability for high-resolution voxel modeling.

Findings

Achieves state-of-the-art results on ShapeNet benchmark.

Demonstrates superior performance in 3D point cloud completion.

Offers faster inference with lower computational costs.

Abstract

Recent advancements in sequence modeling have led to the development of the Mamba architecture, noted for its selective state space approach, offering a promising avenue for efficient long sequence handling. However, its application in 3D shape generation, particularly at high resolutions, remains underexplored. Traditional diffusion transformers (DiT) with self-attention mechanisms, despite their potential, face scalability challenges due to the cubic complexity of attention operations as input length increases. This complexity becomes a significant hurdle when dealing with high-resolution voxel sizes. To address this challenge, we introduce a novel diffusion architecture tailored for 3D point clouds generation-Diffusion Mamba (DiM-3D). This architecture forgoes traditional attention mechanisms, instead utilizing the inherent efficiency of the Mamba architecture to maintain linear complexity with respect to sequence length. DiM-3D is characterized by fast inference times and substantially lower computational demands, quantified in reduced Gflops, thereby addressing the key scalability issues of prior models. Our empirical results on the ShapeNet benchmark demonstrate that DiM-3D achieves state-of-the-art performance in generating high-fidelity and diverse 3D shapes. Additionally, DiM-3D shows superior capabilities in tasks like 3D point cloud completion. This not only proves the model's scalability but also underscores its efficiency in generating detailed, high-resolution voxels necessary for advanced 3D shape modeling, particularly excelling in environments requiring high-resolution voxel sizes. Through these findings, we illustrate the exceptional scalability and efficiency of the Diffusion Mamba framework in 3D shape generation, setting a new standard for the field and paving the way for future explorations in high-resolution 3D modeling technologies.