PaperScope
LIVE · 2026-09-29 05:40 UTC

Proxy2World: Learning to Generate Worlds From Lightweight Proxies without Seeing Them

Hongli Xu, Weilong Yan, Anbang Wang, Chunyu Zou, Siyu Hong, Jingwei Huang

Latestcs.CLcs.LGcs.AIcs.CV
arXiv ID
2609.35023 v1
Category
Submitted
2026-09-28

Abstract

Lightweight scene proxies let creators control scene layout and motion while leaving room for imagination in appearance, lighting, and visual effects. However, a suitable proxy is not uniquely defined, making paired proxy-video data difficult to construct automatically at scale. We present Proxy2World, a controllable world model that learns these complementary capabilities from ordinary posed RGBD videos, without training on authored proxy-video pairs. The model jointly learns depth-conditioned RGB generation and joint RGBD generation through cross-modal flow matching. Learning both tasks enables proxy-camera hybrid denoising at inference to follow the proxy structure while producing natural, detailed visuals. We further introduce ProxyBench to evaluate this capability across a diverse set of scenes, camera trajectories, and subject motions. Experiments on ProxyBench show that Proxy2World achieves a better balance between structural adherence and visual quality than camera-controlled and geometry-conditioned methods, supported by quantitative metrics, VLM assessments, human evaluations and diverse qualitative results.

Comment: Project page: https://dumdumgura.github.io/proxy2world/

arXiv abs page · PDF · same-day batch