Clutter Pick — Unitree G1 Humanoid Teleoperation: "Pick a Target Object from a Cluttered Bin" Demonstrations (LeRobot)

Multi-view humanoid teleoperation demonstrations of a cluttered-bin pick-and-place manipulation task, captured on a Unitree G1 robot operated by a human teleoperator. Task: "Pick a target object from the cluttered bin and place it at the target location." Packaged as a HuggingFace LeRobot v2.1 dataset (3 episodes, 5,142 frames total) and QC-complete (Tier-1, 0 FAILs). Six video streams at ~30 fps, all H.264: head RGB (1920×1080), left + right color RGB (1280×720), a head-depth visualization (1280×720), a LiDAR render (640×480), and a third-person scene overview (1280×548). The parquet carries dense proprioception — 26-DOF joint state, 29-DOF action, joint velocity/torque, Cartesian pose/velocity/wrench, fingertip positions, tactile (normal/tangential/proximity), and a 4,096-point LiDAR point cloud per frame — plus frame timestamps + indices. Useful for cluttered-scene grasping, pick-and-place policy learning, multi-view + depth-aware representation learning, and VLA pretraining on humanoid teleoperation data.