Ropedia Xperience 2: A Gold Standard for Scaling Physical AI Environments and Behaviors
Abstract
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.
Figures & tables
| Device | Form factor | Cameras | Shutter | LiDAR | IMU | Audio channels |
| 3D scanner | Handheld | 2 fisheye, 2 wide-angle | Rolling | ✓ | — | — |
| HOMIE | Head-mounted | 4 fisheye (2 front, 2 rear) | Global | — | ✓ | 4 |
| Asset | Device | Frame | Rate | Provenance |
| Fisheye images | 3D scanner | Scene | Per scan | Observed |
| Camera poses, mesh, 3DGS | 3D scanner | Scene | Per scan | Estimated |
| Fisheye video ( 4), audio, IMU | HOMIE | Device | 30 fps / 200 Hz | Observed |
| Device pose, sparse map | HOMIE | Recording world | 30 Hz | Estimated |
| Stereo depth, confidence | HOMIE | Stereo-left camera | 30 Hz | Estimated |
| Scene registration | Both | Recording scene | Per recording | Estimated |
| Unit | Definition | Count | Breakdown |
| Location | A home or another independent capture site | 596 | Figure 7 , right |
| Scene (Level 1) | One released 3D scan and its reconstruction | 1,589 | Figures 6 and 7 |
| Task (Level 2) | A task definition with a goal and instructions | 74 | Figure 4 a–b |
| Recording | One uploaded HOMIE recording of one task | 4,230 | Figure 4 c |
| Duration | Total HOMIE collection time | 131 h | — |
| Sparse (5 frames) | Medium (11–100 frames) | Dense (335–500 frames) | |||||||
| Model | AbsRel | RMSE | AbsRel | RMSE | AbsRel | RMSE | |||
| Optimization-based | |||||||||
| DUSt3R [ 57 ] | 0.1485 | 0.6618 | 0.3128 | 0.1758 | 0.6478 | 0.1970 | |||
| MASt3R † [ 24 ] | 0.1174 | 0.5946 | 0.2983 | 0.1737 | 0.6264 | 0.2406 | |||
| End-to-end feed-forward | |||||||||
| VGGT [ 55 ] | 0.0833 | 0.4554 | 0.3136 | 0.1043 | 0.4425 | 0.2435 | 0.1071 | 0.2718 | 0.1978 |
| End-to-end models | 3DGS fields | Lifting | |||||
| Metric | SpatialLM [ 34 ] | LLaVA-3D ∗ [ 64 ] | SceneScript [ 2 ] | Map-Det3D [ 60 ] | M3 [ 67 ] | LangSplat [ 44 ] | Boxer [ 13 ] |
| AP | 4.23 | 0.00 | 0.00 | 6.45 | 5.72 | 3.43 | 24.69 |
| AP | 1.07 | 0.00 | 0.00 | 0.13 | 0.64 | 0.34 | 6.09 |
| [email protected] | 43.66 | 0.00 | 0.00 | 36.16 | 37.01 | 37.96 | 49.48 |
| Model | Trans. | Rot. | Dist. | Depth | Height | Rel.Pos. | Overall |
| Open-weight models | |||||||
| Qwen2.5-VL-7B [ 3 ] | 20.24 | 36.14 | 17.86 | 65.06 | 79.52 | 78.31 | 49.40 |
| Qwen2.5-VL-72B [ 3 ] | 21.43 | 48.19 | 28.57 | 81.93 | 92.77 | 90.36 | 60.40 |
| InternVL3-8B [ 65 ] | 25.00 | 30.12 | 19.05 | 68.67 | 95.18 | 80.72 | 53.00 |
| InternVL3-78B [ 65 ] | 36.90 | 46.99 | 14.29 | 80.72 | 93.98 | 84.34 | 59.40 |
| LLaVA-OneVision-7B [ 25 ] | 15.48 | 24.10 | 13.10 | 56.63 | 85.54 | 74.70 | 44.80 |
Appendix figures & tables3 assets
Supplementary material from the paper’s appendix.
Appendix
| Category | Perfect | Usable | Reuse | Released | Delete | Collection sites |
| Home | ||||||
| Living room | 19 | 220 | 100 | 339 | 0 | 443 |
| Bedroom | 60 | 434 | 143 | 637 | 8 | |
| Bathroom | 3 | 39 | 58 | 100 | 9 | |
| Kitchen | 1 | 13 | 62 | 76 | 3 | |
| Multi-room | 6 | 133 | 98 | 237 | 0 | |
| Environment | Perfect | Usable | Total |
| Indoor | 8 | 7 | 15 |
| Outdoor | 5 | 0 | 5 |
| All | 13 | 7 | 20 |
| Track | Recorded settings |
| All tracks | Model/checkpoint or API version; code revision; resolution; eligible scenes; failures; hardware. |
| Depth and geometry | Sparse/Medium/Dense setting; selected frame IDs, count, and order; supplied priors; depth alignment; valid coverage. |
| 3D object detection | Query categories; input images, point cloud, or 3DGS; 2D proposals; box parameterization. |
| Spatial intelligence | Input views; prompt; decoding settings; answer extraction; question counts by type. |
| Virtual teleoperation | Converted kitchen scene; head-mounted display and hand input devices; kettle-transfer interaction; robot and external camera views. |
Robotic Data Collection
Robotic data collection develops ways to gather, label, and validate robot and human-generated experience for learning, evaluation, and sensing.
See all 196 papers