ObjectWorld: Scaling Object-Centric World Models with A Large-Scale 3D Full-Object Flow Dataset
Overview
This paper proposes 3D full-object flow, an object-centric motion representation that assigns dense 3D trajectories to points across the complete surface of a target object, including occluded regions. The authors introduce a 1210-hour full-object flow dataset integrating diverse real-world videos and physical simulations with a scalable annotation pipeline. A world model is trained to predict this flow from images and task descriptions, demonstrating reliable annotation quality, data-scalable flow prediction, and effective downstream manipulation across novel scenes and robot embodiments.
Generated from attributed reports · Updated 11 hours ago
Event evidence and corrections
0 attributed source owners. Ownership does not establish independent confirmation. Quantities are reported separately and are never added together.
Reported quantity · Reported duration: 1210 hours · Basis not reported
“lete surface of a target object, including its occluded regions. By extending visible-surface object flow to the full object geometry, this representation makes complete object motion explicit and applies uniformly to both rigid and deformable objects. To realize the potential of this representation at scale, we introduce a 1210-hour full-object flow dataset that addresses the scarcity of large-scale supervision for”
Exact source · revision 1Source owner not reported
Report timeline
Follow attributed reports and material updates.
- arXiv Robotics — research abstractsSignalObjectWorld: Scaling Object-Centric World Models with A Large-Scale 3D Full-Object Flow Dataset
This paper proposes 3D full-object flow, an object-centric motion representation that assigns dense 3D trajectories to points across the complete surface of a target object, including occluded regions. The authors introduce a 1210-hour full-object flow dataset integrating diverse real-world videos and physical simulations with a scalable annotation pipeline. A world model is trained to predict this flow from images and task descriptions, demonstrating reliable annotation quality, data-scalable flow prediction, and effective downstream manipulation across novel scenes and robot embodiments.
Event coverage history
There is not enough continuous observation data to show a trend.
Timezone · UTC
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).