Tutorials
Guiding Image-to-3D Generation with Test-Time Partial Observations
arXiv:2609.10531v1 Announce Type: new Abstract: Image-to-3D models can generate visually compelling 3D assets from a single RGB image, but their geometry is often only loosely constrained by the avail
arXiv:2609.10531v1 Announce Type: new Abstract: Image-to-3D models can generate visually compelling 3D assets from a single RGB image, but their geometry is often only loosely constrained by the available observations, limiting their use in applications that require geometric fidelity. In many real-world settings, however, partial geometric observations of the object may be available at test time. We introduce a training-free framework for incorporating such evidence into pretrained image-to-3D generative models without retraining or finetuning. To do this, we guide generation using a ray-consistent observation likelihood defined over the model's occupancy representation, combining surface occupancy and free-space evidence. Applied to SAM 3D and its multi-view extension, our approach substantially improves geometric fidelity across different levels of observability, as well as visual quality. Our results demonstrate that pretrained image-to-3D models can effectively integrate partial geometric observations through explicit test-time guidance, complementing their learned generative priors without modifying the underlying model.
Related
- Realiz3D: 3D Generation Made Photorealistic via Domain-Aware Learning
- Point-MF: One-step Point Cloud Generation from a Single Image via Mean Flows
- CORGI: Consistency-Aware 3D Dog Reconstruction from a Single Image in the Wild
- Geometry Matters: 3D Foundation Priors for Learning Semantic Correspondence
Source: arXiv cs.CV | 2026-09-10