Datasetpublic

brandonyang/dual-lidar-umi-independent

This dataset was created using LeRobot. Dataset Description Standalone LeRobot v3 dataset containing 296 dual-UMI orange-collection demonstrations (276,332 frames, 2.559 hours at 30 FPS). It contains synchronized observation.images.umi1 and observation.images.umi2 video observations, 14D observation.state, and 14D action; no LiDAR files or LiDAR frame features are included. For each UMI independently, pose is expressed relative to that UMI's episode-start pose using… See the full description on the dataset page: https://huggingface.co/datasets/brandonyang/dual-lidar-umi-independent.

sourceHugging Faceapache-2.0updated 2mo agoView on Hugging Face

0likes156downloads

Dataset Card

This dataset was created using LeRobot.

Dataset Description

Standalone LeRobot v3 dataset containing 296 dual-UMI orange-collection demonstrations (276,332 frames, 2.559 hours at 30 FPS). It contains synchronized observation.images.umi1 and observation.images.umi2 video observations, 14D observation.state, and 14D action; no LiDAR files or LiDAR frame features are included.

For each UMI independently, pose is expressed relative to that UMI's episode-start pose using T_relative(t) = inverse(T_start) @ T(t). Each 7D UMI state is [x, y, z, rotation-vector-x, rotation-vector-y, rotation-vector-z, normalized-gripper]. Grippers use 0=closed, 1=open, with physical ranges of 56-125 mm for UMI1 and 68-119 mm for UMI2. action[t] is intentionally an exact same-frame copy of observation.state[t]; training code may derive a future action representation.

The sole task is pick up all the oranges and put them in the bowl. The train split contains all 296 episodes. Sources and immutable revisions are brandonyang/dual-lidar-umi@f54456b93896ea0f9c6c81b8bc54c96a67828b0b and brandonyang/dual-lidar-umi-relative@3745003600a99b5e67cc2adfed11c53edce982a1. Per-episode provenance and filtering diagnostics are available under meta/.

—Homepage: [More Information Needed]
—Paper: [More Information Needed]
—License: apache-2.0

Dataset Structure

meta/info.json:

json

{
    "codebase_version": "v3.0",
    "fps": 30,
    "features": {
        "observation.state": {
            "dtype": "float32",
            "shape": [
                14
            ],
            "names": [
                "umi1_x",
                "umi1_y",
                "umi1_z",
                "umi1_rx",
                "umi1_ry",
                "umi1_rz",
                "umi1_gripper",
                "umi2_x",
                "umi2_y",
                "umi2_z",
                "umi2_rx",
                "umi2_ry",
                "umi2_rz",
                "umi2_gripper"
            ]
        },
        "action": {
            "dtype": "float32",
            "shape": [
                14
            ],
            "names": [
                "umi1_x",
                "umi1_y",
                "umi1_z",
                "umi1_rx",
                "umi1_ry",
                "umi1_rz",
                "umi1_gripper",
                "umi2_x",
                "umi2_y",
                "umi2_z",
                "umi2_rx",
                "umi2_ry",
                "umi2_rz",
                "umi2_gripper"
            ]
        },
        "observation.images.umi1": {
            "dtype": "video",
            "shape": [
                600,
                800,
                3
            ],
            "names": [
                "height",
                "width",
                "channels"
            ],
            "info": {
                "video.height": 600,
                "video.width": 800,
                "video.codec": "h264",
                "video.pix_fmt": "yuv420p",
                "video.fps": 30,
                "video.channels": 3,
                "has_audio": false,
                "video.g": 2,
                "video.crf": 30,
                "video.preset": null,
                "video.fast_decode": 0,
                "video.video_backend": "pyav",
                "video.extra_options": {},
                "is_depth_map": false
            }
        },
        "observation.images.umi2": {
            "dtype": "video",
            "shape": [
                600,
                800,
                3
            ],
            "names": [
                "height",
                "width",
                "channels"
            ],
            "info": {
                "video.height": 600,
                "video.width": 800,
                "video.codec": "h264",
                "video.pix_fmt": "yuv420p",
                "video.fps": 30,
                "video.channels": 3,
                "has_audio": false,
                "video.g": 2,
                "video.crf": 30,
                "video.preset": null,
                "video.fast_decode": 0,
                "video.video_backend": "pyav",
                "video.extra_options": {},
                "is_depth_map": false
            }
        },
        "timestamp": {
            "dtype": "float32",
            "shape": [
                1
            ],
            "names": null
        },
        "frame_index": {
            "dtype": "int64",
            "shape": [
                1
            ],
            "names": null
        },
        "episode_index": {
            "dtype": "int64",
            "shape": [
                1
            ],
            "names": null
        },
        "index": {
            "dtype": "int64",
            "shape": [
                1
            ],
            "names": null
        },
        "task_index": {
            "dtype": "int64",
            "shape": [
                1
            ],
            "names": null
        }
    },
    "total_episodes": 296,
    "total_frames": 276332,
    "total_tasks": 1,
    "chunks_size": 1000,
    "data_files_size_in_mb": 100,
    "video_files_size_in_mb": 200,
    "data_path": "data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet",
    "video_path": "videos/{video_key}/chunk-{chunk_index:03d}/file-{file_index:03d}.mp4",
    "robot_type": "dual_umi",
    "splits": {
        "train": "0:296"
    }
}

Citation

BibTeX:

bibtex

[More Information Needed]