Unassigned · 2026-11-13
Ropedia Xperience 2: A Gold Standard for Scaling Physical AI Environments and Behaviors
Hao Li, Fangzhou Hong, Zhaoxi Chen, Hui En Pang, Kairui Hu, Mingze Sun, Runmao Yao, Xinhao Ji, Xiyuan Zheng, Xuanli Chen, Yalun Dai, Yap Kim Hui, Dingwen Zhang, Yuanyuan Gao, Ziang Cao, Cheng Huang Yean, Jackson Peng, Rui Duan, Yaobo Liu, Zetong Wang, Ziwei Liu
AI-assisted summary based on the paper. Consult the linked source for methods and limitations. Benchmark numbers appear after editor verification.
TL;DR
The authors introduce Ropedia Xperience 2, a physical-AI dataset combining handheld LiDAR 3D scans (1,589 released metric-scale scenes) with 4,230 egocentric HOMIE task recordings covering 74 tasks (131 hours). The captured scenes can be converted to physically interactive simulation, human motions can be retargeted to whole-body robot trajectories, and additional robot behaviors can be collected via virtual teleoperation and long-horizon trajectory generation.
Why it matters
The paper targets brittleness of direct visual robot control on long-horizon contact-rich real tasks, and argues that scaling physical AI needs both environment data and behavior demonstrations. Ropedia Xperience 2 provides scan-based references for depth/geometry/object/spatial-reasoning benchmarks and a pipeline to reuse the same captures for robot learning via real-to-sim, motion retargeting, virtual teleoperation, and automated trajectory generation.
Method
- Collect 1,589 released metric-scale 3D scans using a handheld LiDAR scanner, and 4,230 egocentric recordings (HOMIE) of 74 tasks (131 hours) with a protocol requiring no robots or motion-capture infrastructure on site.
- Process scans into meshes and 3D Gaussian splatting; process HOMIE into synchronized video, device trajectories, hand/body motion (MANO, SMPL-H), and hierarchical language annotations.
- Register HOMIE recordings to scan geometry to obtain a shared metric scene frame, then reuse captures for real-to-sim conversion, human-robot retargeting, virtual teleoperation, and long-horizon trajectory generation.
Limitation
A single transform cannot correct local SLAM drift within a recording; per-segment correction is left to future work.
Abstract (from arXiv)
Large foundation models such as GPT-6 Astra and Claude Fable 5.1 have recently been shown to control robots directly from visual observations. They nevertheless remain brittle on long-horizon, contact-rich tasks in real-world environments. Further progress depends on two complementary resources: demonstrations of human behavior that supervise how tasks are performed, and interactive environments in which policies can be rolled out, evaluated, and improved in closed loop. Language models have advanced along well-characterized scaling laws; the corresponding question for physical AI is how capability scales with environments and behaviors, which first requires identifying which data to scale and how to produce and reuse such data continuously. To address this question, we introduce Ropedia Xperience 2, comprising 1,589 released metric-scale 3D scans and a collection of 4,230 egocentric recordings of 74 tasks (131 hours). We further show that each capture can be repurposed for robot learning: scenes are converted into physically interactive simulation environments, human motion is retargeted to whole-body robot trajectories, and additional robot behaviors are obtained through virtual teleoperation and automated long-horizon trajectory generation. We propose this joint capture of environments and behaviors as a gold standard for physical AI data collection, laying a data foundation for establishing scaling laws over environments and behaviors.
Related papers
- Ropedia W0: Scaling Loco-Manipulation World Models across Embodiments with Human Experience
- Open-HumanWAM: An Open Recipe for Scaling Human and Robot Experience for Embodied Foundation Models
- RobotWorld: Benchmarking Multimodal Agents for Robot Use Across Diverse Tasks and Embodiments
- Gym-Anything for Robotics: Scaling physical simulations with digital agents
- Real2Gym: Building Gyms from Videos, Bringing Skills to Robots
Publication record
Bibliographic evidence describes where and when a record appeared. It does not validate scientific claims.
First public release: 2026-11-12 · Indexed by Pub-AI: 2026-11-13 UTC
Ropedia Xperience 2: A Gold Standard for Scaling Physical AI Environments and Behaviors
Preprint posted: 2026-11-12 · Original record ↗
- arxiv
- 2611.17060
Explicit source versions (1)
- v1 · 2026-11-12
- Preprint available
Source date: 2026-11-12 · First detected: 2026-11-13 UTC · Historical baseline
Source evidence (2)
- Official evidence ↗ · RSS guid=oai:arXiv.org:2611.17060v1; arxiv:announce_type=new; item pubDate (official metadata)
- Official evidence ↗ · atom:published (official metadata)
Abstract · arxiv
Large foundation models such as GPT-6 Astra and Claude Fable 5.1 have recently been shown to control robots directly from visual observations. They nevertheless remain brittle on long-horizon, contact-rich tasks in real-world environments. Further progress depends on two complementary resources: demonstrations of human behavior that supervise how tasks are performed, and interactive environments in which policies can be rolled out, evaluated, and improved in closed loop. Language models have advanced along well-characterized scaling laws; the corresponding question for physical AI is how capability scales with environments and behaviors, which first requires identifying which data to scale and how to produce and reuse such data continuously. To address this question, we introduce Ropedia Xperience 2, comprising 1,589 released metric-scale 3D scans and a collection of 4,230 egocentric recordings of 74 tasks (131 hours). We further show that each capture can be repurposed for robot learning: scenes are converted into physically interactive simulation environments, human motion is retargeted to whole-body robot trajectories, and additional robot behaviors are obtained through virtual teleoperation and automated long-horizon trajectory generation. We propose this joint capture of environments and behaviors as a gold standard for physical AI data collection, laying a data foundation for establishing scaling laws over environments and behaviors.
