ETH Zurich releases EgoHTR: 150k-frame egocentric 4D dataset for humanoid terrain traversal
Scobleizer · x · 2026-11-01
ETH Zurich, with Stanford, UC Berkeley and TU Munich, released EgoHTR, a dataset and open-source reconstruction pipeline targeting humanoid locomotion over unstructured terrain.
- Captured 55 scene-aligned 4D human motion sequences (150k frames at 30fps, 1.37h total, 8 subjects, 7 scenes) using egocentric wearables plus a portable 3D scanner, achieving SOTA accuracy against mocap ground truth
- The pipeline is fully open-source and extensible; perceptive locomotion policies trained on the data were deployed on a Unitree G1 humanoid
- Companion open-source project orcs v0.1.1 demonstrates a single policy handling rocky climbs and window crossing, plus an agentic UMR retargeting skill where an agent revises reference motions from frame-wise notes
- The team will present at CoRL
More from Embodied
- Robot Intelligence Index debuts: GPT-6 Astra tops robot model ranking at 1905 — haider1 · 2026-11-01
- RoboColiseum robot benchmark: Claude Opus 5.5 scores 76.28, edging out GPT-6 Astra — jiqizhixin · 2026-11-01
- Dyna's long-horizon robot demo hailed as one of the cleanest yet — Scobleizer · 2026-11-01
- Tesla Robotaxi out-engages Waymo 9x on X despite smaller fleet — mitchdeg · 2026-11-01
- The Trinity: capability-driven RSI lifts frozen-model benchmarks 34 points, robot success 0→60% — YangYou1991 · 2026-11-01
- 10 spicy takes from 50 frontier robotics insiders: Astra is robotics' GPT-3 moment — ZhitingHu · 2026-11-01