Ropedia Xperience 2: 1,589 3D scans and 131 hours of egocentric data for physical AI
ropedia-ai · hf · 2026-11-13
Ropedia released Xperience 2, proposing a gold standard for physical AI data collection.
- 1,589 metric-scale 3D scans (indoor/outdoor homes, offices, restaurants, stores) plus 4,230 egocentric recordings spanning 74 tasks and 131 hours
- Core thesis: as LLMs advanced via scaling laws, physical AI needs to establish how capability scales with environments and behaviors — which first requires knowing what data to scale and how to produce and reuse it continuously
- Every capture is repurposed for robot learning: scenes become physically interactive simulation environments, human motion is retargeted to whole-body robot trajectories, and additional behaviors come from virtual teleoperation and automated long-horizon trajectory generation
Open-sourced on GitHub and Hugging Face.
Related event: Robopedia releases Xperience 2 physical AI dataset(2 posts)→
More from Embodied
- Robopedia launches Xperience 2, a 'gold standard' Physical AI dataset — liuziwei7 · 2026-11-13
- Baidu invites riders to try its 6th-gen Apollo Go robotaxi in Beijing — Baidu_Inc · 2026-11-13
- Gemini Robotics 2.5 Debuts at CoRL 2026: Humanoid Ties a Bow Autonomously — ZoubinGhahrama1 · 2026-11-13
- LLM edits 65 embedding rows to fix robot failures, lifting held-out success 74.3% to 87.5% — 机器之心 · 2026-11-13
- Astra vs Fable 5.2 zero-shot excavator test: only visual reference, one model prevails — Scobleizer · 2026-11-13
- Waymo rider pulls up next to a Cybercab — with a human driver still in the seat — kscottz · 2026-11-13