paper-with-me

홈 › Papers

GEM-4D: Geometry-Enhanced Video World Models for Robot Manipulation

2026-05-20 · Kaichen Zhou, Yuzhen Chen, Fangneng Zhan, Hang Hua, Grace Chen, Xinhai Chang, Ao Qu, Yilun Du, Zhuang Liu, Paul Pu Liang, Mengyu Wang arxiv

Video world models can generate realistic futures from a single instruction, but they often fail to track the same physical points consistently across time. As a result, the generated videos appear plausible, yet lack the physical grounding required for reliable action execution, such as robot manipulation. We present GEM-4D, a geometry-grounded video world model that resolves this limitation by injecting dense 4D correspondence supervision distilled from a pretrained geometry foundation model into the video generative backbone during training. This supervision enables the model to jointly capture appearance and geometric structure while retaining a single-stream architecture with no additional inference cost. We further introduce an inverse dynamics module that converts correspondence-consistent video rollouts into executable robot trajectories, enabling direct deployment in both real-world and simulated manipulation. GEM-4D achieves state-of-the-art performance on both video prediction and geometric consistency across both simulation and realistic scenarios and improves real-world manipulation success from 61% to 81%. Additional results are available at https://gem-4d.github.io/.

📄 PDF Abstract BibTeX arXiv:2605.22882

Code (0)

등록된 구현이 없습니다.

Tasks

Robot ManipulationVideo Prediction

Similar Papers 제목 키워드 기반

RoboTransfer: Geometry-Consistent Video Diffusion for Robotic Visual Policy Transfer

2025-05-29 · Liu Liu, XiaoFeng Wang, Guosheng Zhao, Keyu Li 외

Imitation Learning has become a fundamental approach in robotic manipulation. However, collecting large-scale real-world robot demonstrations is prohibitively expensive. Simulators offer a cost-effective alternative, but…

Imitation LearningVideo Generation

DynaRend: Learning 3D Dynamics via Masked Future Rendering for Robotic Manipulation

2025-10-28 · Jingyi Tian, Le Wang, Sanping Zhou, Sen Wang 외 arxiv

Learning generalizable robotic manipulation policies remains a key challenge due to the scarcity of diverse real-world training data. While recent approaches have attempted to mitigate this through self-supervised repres…

Representation LearningVideo Prediction

Robotic Manipulation is Vision-to-Geometry Mapping ($f(v) \rightarrow G$): Vision-Geometry Backbones over Language and Video Models

2026-04-14 · Zijian Song, Qichang Li, Jiawei Zhou, Zhenlong Yuan 외 arxiv

At its core, robotic manipulation is a problem of vision-to-geometry mapping ($f(v) \rightarrow G$). Physical actions are fundamentally defined by geometric properties like 3D positions and spatial relationships. Consequ…

Zero-shot Generalization

GenVid2Robot: From Video Generation to Robot Manipulation via Rigid-Geometric Consistency

2026-07-10 · Haohui Huang, Xi Yuan, Panpan Liao, Tao Teng 외 arxiv

Generated videos provide useful visual motion priors for robot manipulation, but their visual plausibility does not imply physical executability. A generated video usually lacks metric geometry, grasp grounding, robot ki…

Robot ManipulationVideo Generation

Learning 4D Geometric Priors for Inference-Efficient World Action Models

2026-07-06 · Jianjun Zhang, Jian Zhu, Taiyi Su, Chong Ma 외 arxiv

World Action Models (WAMs) have shown strong potential for robotic manipulation by jointly modeling visual future dynamics and executable action sequences. However, existing video-action co-training methods primarily opt…