HierOctFusion: Multi-scale Octree-based 3D Shape Generation via Part-Whole-Hierarchy Message Passing
This work addresses the challenge of generating complex 3D objects more effectively for applications in computer graphics and AI, though it appears incremental by building on existing octree-based approaches.
The paper tackled the problem of 3D shape generation by addressing limitations in existing octree-based diffusion models, such as ignoring semantic part hierarchies and computational inefficiency, and proposed HierOctFusion, which achieved superior shape quality and efficiency compared to prior methods.
3D content generation remains a fundamental yet challenging task due to the inherent structural complexity of 3D data. While recent octree-based diffusion models offer a promising balance between efficiency and quality through hierarchical generation, they often overlook two key insights: 1) existing methods typically model 3D objects as holistic entities, ignoring their semantic part hierarchies and limiting generalization; and 2) holistic high-resolution modeling is computationally expensive, whereas real-world objects are inherently sparse and hierarchical, making them well-suited for layered generation. Motivated by these observations, we propose HierOctFusion, a part-aware multi-scale octree diffusion model that enhances hierarchical feature interaction for generating fine-grained and sparse object structures. Furthermore, we introduce a cross-attention conditioning mechanism that injects part-level information into the generation process, enabling semantic features to propagate effectively across hierarchical levels from parts to the whole. Additionally, we construct a 3D dataset with part category annotations using a pre-trained segmentation model to facilitate training and evaluation. Experiments demonstrate that HierOctFusion achieves superior shape quality and efficiency compared to prior methods.