arXiv cs.AI· Yeobin Hong, Hyojun Go, Dohun Lee, Jeongjae Lee, Sunghwan Hong, Federico Tombari, Zhaochong An, Jong Chul Ye, Konrad Schindler, Dominik Narnhofer·· 3 天前AI 评分27
GeoFix：用单一几何奖励修复 4D 生成一致性
GeoFix: Geometry Fixing for 4D Consistent Generation
AI 导读
GeoFix 是一个统一框架，仅用单一几何奖励即可提升 4D 生成的几何一致性与视觉质量，适用于训练和推理阶段的奖励对齐。该奖励通过校准相机视场重叠带来的偏差，衡量静态场景区域的跨视角深度一致性，避免约束外观或惩罚合理物体运动。结果显示，WorldScore 3D 一致性提升 39%，VBench Quality Score 提升 4%。
正文 · 原文
Abstract:Video diffusion models increasingly serve as priors for 4D generation, but the generated videos often lack geometric consistency. Recent approaches address this through post-training by designing geometric rewards, yet this can degrade texture and fine detail, requiring auxiliary appearance rewards and careful reward balancing. To address this challenge, we propose GeoFix, a unified framework that improves geometric consistency and visual quality with a single geometric reward, applicable to both training and inference-time reward alignment. Our reward measures cross-view depth consistency in static scene regions, targeting structural inconsistencies without directly constraining appearance or penalizing valid object motion. Crucially, depth agreement increases with camera field-of-view overlap even in real videos and 4D datasets, allowing a smaller camera movement to be mistaken for better geometry. We account for this effect by calibrating the reward against the overlap-dependent trend. GeoFix achieves a 39% improvement in WorldScore 3D consistency and a 4% increase in VBench Quality Score.
| Comments: | Project Page: this https URL |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.14293 [cs.CV] |
| (or arXiv:2610.14293v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.14293 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yeobin Hong [view email]
[v1]
Fri, 9 Oct 2026 20:15:30 UTC (3,488 KB)
来源：arXiv cs.AI · arxiv.org