Paper Overview
Field: Computer Vision (CV) Authors: Jerred Chen, Simon Weber, Ronald Clark Published: 2026-09-09 arXiv: 2609.10531
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.
The authors introduce a training-free framework for incorporating such evidence into pretrained image-to-3D generative models without retraining or finetuning. To do this, they 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, the approach substantially improves geometric fidelity across different levels of observability, as well as visual quality. The results demonstrate that pretrained image-to-3D models can effectively integrate partial geometric observations through explicit test-time guidance, complementing their learned generative prior without modifying the underlying model.
Key Contributions
- Training-free guidance: Integrates partial geometric evidence into pretrained image-to-3D models at test time, with no retraining or finetuning required.
- Ray-consistent observation likelihood: Defined over the model's occupancy representation, combining both surface occupancy and free-space evidence.
- Validated on SAM 3D: Substantial improvements in geometric fidelity and visual quality across different levels of observability.