| title | GATOR: Generative and Agentic 3D Object Reconstruction From Casual Images |
|---|---|
| authors | Qirui Wu, Stan Birchfield, Hesam Rabeti, Angel X. Chang, Bowen Wen |
| date | 2026-10-08 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2610.11215 |
| affiliation | NVIDIA |
| upvotes | 2 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf |
| abstract | Reconstructing complete, scene-aligned 3D objects from casual images requires integrating sparse, uncertain observations and inferring surfaces hidden by occlusions. We present GATOR, a generative and agentic framework that recovers textured object assets and their scene-relative pose from one or more images. Our local modality mixer couples patch-aligned RGB, target-mask, and pointmap features before cross-view reasoning, preserving scene context while distinguishing the target from its surroundings. Text-guided semantic conditioning complements these spatial cues with category names and object captions through stage-specific adapters for structure, geometry, and appearance generation. The generated asset initializes a multimodal agent, providing instance-specific geometry and pose for targeted structural and texture refinement through an observation-guided edit-render-review loop. Across synthetic objects, cluttered tabletops, and indoor scenes, GATOR achieves strong geometric and appearance fidelity while recovering scene-relative pose from sparse observations. Time-budget comparisons and scene-level simulation further demonstrate the reconstruction efficiency and simulation readiness. Project page: https://research.nvidia.com/labs/lpr/gator/ |
cite
@misc{wu2026gator,
title={GATOR: Generative and Agentic 3D Object Reconstruction From Casual Images},
author={Qirui Wu and Stan Birchfield and Hesam Rabeti and Angel X. Chang and Bowen Wen},
year={2026},
eprint={2610.11215},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2610.11215}
}