3D Neural Field Generation using Triplane Diffusion
This provides an efficient method for 3D content generation, which is incremental as it adapts existing 2D diffusion models to 3D tasks.
The paper tackles 3D-aware generation of neural fields by converting 3D data into 2D triplane feature representations and training existing 2D diffusion models on them, achieving state-of-the-art results with high quality and diversity on ShapeNet object classes.
Diffusion models have emerged as the state-of-the-art for image generation, among other tasks. Here, we present an efficient diffusion-based model for 3D-aware generation of neural fields. Our approach pre-processes training data, such as ShapeNet meshes, by converting them to continuous occupancy fields and factoring them into a set of axis-aligned triplane feature representations. Thus, our 3D training scenes are all represented by 2D feature planes, and we can directly train existing 2D diffusion models on these representations to generate 3D neural fields with high quality and diversity, outperforming alternative approaches to 3D-aware generation. Our approach requires essential modifications to existing triplane factorization pipelines to make the resulting features easy to learn for the diffusion model. We demonstrate state-of-the-art results on 3D generation on several object classes from ShapeNet.