| title | GenRecon: Bridging Generative Priors for Multi-View 3D Scene Reconstruction |
|---|---|
| authors | Katharina Schmid, Nicolas von Lützow, Jozef Hladký, Angela Dai, Matthias Nießner |
| date | 2026-05-22 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2605.23888 |
| upvotes | 14 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf |
| abstract | We introduce a new approach to high-fidelity 3D scene reconstruction from multi-view RGB images that tightly couples reconstruction with a strong generative 3D prior. We cast scene reconstruction as conditional 3D generation over a set of spatially-localized, overlapping chunks that together tile the scene, scaling generation to large scene extents. Crucially, we inherit the fidelity and completeness of state-of-the-art generative shape models -- we use Trellis.2 as an example -- which we generalize to the scene level. To this end, we propose a projection-based conditioning mechanism that lifts posed multi-view image features into a coherent 3D representation aligned with the generative model, independent of view ordering and spatially anchored to the scene, yielding high-fidelity, multi-view consistent generated geometry. This enables lifting the strong object-level prior of Trellis.2 to multi-view, scene-scale generation, producing faithful, editable PBR mesh reconstructions of indoor environments. As a result, we obtain high-fidelity results that outperform cutting-edge reconstruction methods by 16%. |
cite
@misc{schmid2026genrecon,
title={GenRecon: Bridging Generative Priors for Multi-View 3D Scene Reconstruction},
author={Katharina Schmid and Nicolas von Lützow and Jozef Hladký and Angela Dai and Matthias Nießner},
year={2026},
eprint={2605.23888},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2605.23888}
}