datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
ROPE
Dataset Card for ROPE
The dataset used in this study is designed to evaluate and analyze multi-object hallucination by leveraging existing panoptic segmentation datasets. Specifically, it includes data from MSCOCO-Panoptic and ADE20K, ensuring access to diverse objects and their instance-level semantic annotations.
For more information, please visit Multi-Object Hallucination.
Dataset Construction
The dataset is divided into several subsets based on the distribution… See the full description on the dataset page: https://huggingface.co/datasets/sled-umich/ROPE.MindCraftMindCraft2umi-okra-dex1
UMI Okra Grasping Dataset — Lab Subset (Unitree Dex1-1)
Universal Manipulation Interface (UMI) hand-held teleoperation data for an okra-fruit
grasping / harvesting task. Recorded to train a Diffusion Policy (with a comparison
ACT track) deployed on a Unitree G1.
This is the indoor lab subset. Every session here was recorded in a lab mock okra
field (artificial foliage, white-walled room). The outdoor sessions from the original
collection are not included — see Scope and… See the full description on the dataset page: https://huggingface.co/datasets/Orboh/umi-okra-dex1.umie_datasets
Dataset Card for Dataset Name
UMIE (Unified Medical Imaging Ensemble) is currently the largest publicly available dataset of annotated radiological imaging, combining over 20 open-source datasets into a unified collection with standardized formatting and labeling based on the RadLex ontology.
Dataset Details
Dataset Description
UMIE datasets combine more than 20 open-source medical imaging datasets, containing over 1 million radiological images across… See the full description on the dataset page: https://huggingface.co/datasets/lion-ai/umie_datasets.real01b-routing-d1-blind-dagger-r9-umirel-ef50cov50This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"fps": 15,
"features": {
"observation.state": {
"dtype": "float32",
"shape": [
7
],
"names": [
"cart_pos_x",
"cart_pos_y",
"cart_pos_z",
"cart_rot_x",
"cart_rot_y",
"cart_rot_z"… See the full description on the dataset page: https://huggingface.co/datasets/ankile/real01b-routing-d1-blind-dagger-r9-umirel-ef50cov50.umi_sim_green_cube_v3This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.1",
"robot_type": "xarm6",
"total_episodes": 100,
"total_frames": 24412,
"total_tasks": 1,
"total_videos": 0,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 30,
"splits": {
"train": "0:100"
},
"data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/gellert5225/umi_sim_green_cube_v3.umi_sim_green_cube_v2This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.1",
"robot_type": "xarm6",
"total_episodes": 50,
"total_frames": 13399,
"total_tasks": 1,
"total_videos": 0,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 30,
"splits": {
"train": "0:50"
},
"data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/gellert5225/umi_sim_green_cube_v2.UMI-VQA-8M
VISTA: Vision-Grounded and Physics-Validated Adaptation of UMI Data for VLA Training
🔭 Overview
UMI-VQA-8M is a large-scale visual question answering dataset built for UMI-style wrist-mounted fisheye observations. It contains 8 M visual question-answering samples and provides visual-language supervision for UMI observation scenarios.
The dataset consists of two complementary parts:
Real-World UMI VQAThis subset contains 3 M real-world… See the full description on the dataset page: https://huggingface.co/datasets/TeleEmbodied/UMI-VQA-8M.umi_sim_green_cube_v1This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.1",
"robot_type": "xarm6",
"total_episodes": 50,
"total_frames": 13802,
"total_tasks": 1,
"total_videos": 0,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 30,
"splits": {
"train": "0:50"
},
"data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/gellert5225/umi_sim_green_cube_v1.VL-Illusion
VL Illusion Dataset
Paper:
Homepage: vl-illusion.github.io
drill_play_franka_franka_umi_tacticleThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": null,
"total_episodes": 32,
"total_frames": 16000,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,"video_files_size_in_mb": 200,
"fps": 10,
"splits": {
"train": "0:32"
},
"data_path": "data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/Helium-bre/drill_play_franka_franka_umi_tacticle.umi-okra-lab-20260723
UMI Okra Harvesting — Lab Indoor Session (2026-07-23)
Unitree Dex1-1 グリッパに改造した UMI (Universal Manipulation Interface)
で、研究室内でオクラ収穫動作を撮影したデモンストレーションデータ。
このセッションは 8 本撮影して 8 本すべてが採用された（歩留まり 100%、生データの 99% を使用）。
Orboh が取得した全 6 セッション中で最もクリーンな結果。
English summary: Human demonstrations of okra-harvesting motions recorded in a lab room with a
hand-held UMI gripper (modified with a Unitree Dex1-1) and a GoPro. All 8 recorded demos passed
SLAM relocalization and quality filters. Contains a… See the full description on the dataset page: https://huggingface.co/datasets/Kota0612/umi-okra-lab-20260723.umi-okra-indoor-20260804
UMI Okra Harvesting — Indoor Session (2026-08-04)
Unitree Dex1-1 グリッパに改造した UMI (Universal Manipulation Interface)
で、屋内に並べた観葉植物を対象にオクラ収穫動作を撮影したデモンストレーションデータ。
English summary: Human demonstrations of okra-harvesting motions recorded indoors with a
hand-held UMI gripper (modified with a Unitree Dex1-1) and a GoPro. Contains a
diffusion-policy-ready replay buffer (14 episodes / 3,500 frames) plus the raw ORB-SLAM3
trajectories and calibration files. Read the Limitations section… See the full description on the dataset page: https://huggingface.co/datasets/Kota0612/umi-okra-indoor-20260804.umi_sim_sort_v2This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.1",
"robot_type": "xarm6",
"total_episodes": 46,
"total_frames": 31020,
"total_tasks": 1,
"total_videos": 0,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 30,
"splits": {
"train": "0:46"
},
"data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/gellert5225/umi_sim_sort_v2.dex1-1-umi-fiducial-cube-20260902
Dex1-1 Fiducial Cube Dataset (2026-09-02)
UMI (Universal Manipulation Interface) 形式のデータセットです。Fiducial Cube +
GoProを使ったハンドヘルドデータ収集で取得した、Dex1-1カッターの軌道データです。
収集パイプラインは RoboHarvest
（Autonomous Robotic Pepper Harvesting）を
Dex1-1カッター向けに改修したものです。
内容
154エピソード、36,440フレーム（俯瞰カメラでのFiducial Cube検出 + GoPro IMU
をEKFで融合して軌道を復元）
観測: camera0_rgb（gripper側カメラ映像, 224x224）
行動: エンドエフェクタ位置 eef_pos(3次元) + 姿勢 eef_rot_axis_angle(3次元)
の計6次元。gripper開閉幅は含みません（Dex1-1カッターは開閉幅を記録せず、… See the full description on the dataset page: https://huggingface.co/datasets/Kota0612/dex1-1-umi-fiducial-cube-20260902.umi-okra-grassland-20260724
UMI Okra Harvesting — Outdoor Grassland Session (2026-07-24)
Unitree Dex1-1 グリッパに改造した UMI (Universal Manipulation Interface)
で、屋外の草地でオクラ収穫動作を撮影したデモンストレーションデータ。
Orboh が取得した 6 セッション中、唯一の屋外・直射日光下で成功したデータ。
同日の別セッション（建物脇の植込み）や後日のオクラ圃場は SLAM が通らなかったが、
草地では 9 本中 8 本のリローカライズに成功している。
English summary: Human demonstrations of okra-harvesting motions recorded outdoors on
grassland with a hand-held UMI gripper (modified with a Unitree Dex1-1) and a GoPro.
This is the only outdoor session in… See the full description on the dataset page: https://huggingface.co/datasets/Kota0612/umi-okra-grassland-20260724.piper-stack-umi-v0This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": null,
"total_episodes": 400,
"total_frames": 222082,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 50,
"splits": {
"train": "0:400"
},
"data_path": "data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/axiboai/piper-stack-umi-v0.UMIE-Visual-QATacRich-Manip-LeRobot-umi-set-the-steel-plate
TacRich-Manip LeRobot v3 — set_the_steel_plate
Multimodal real-robot trajectories for gripper-based contact-rich manipulation,
published in the standard LeRobot v3 layout. This task repository contains
umi/set_the_steel_plate and is private during active collection.
Dataset summary
Property
Value
Repository
qingzhu-robotics/TacRich-Manip-LeRobot-umi-set-the-steel-plate
Collection method
umi
Robot
rm75b-pika-tachin
Episodes
98
Frames
86,392… See the full description on the dataset page: https://huggingface.co/datasets/qingzhu-robotics/TacRich-Manip-LeRobot-umi-set-the-steel-plate.COMFORTrobosuite_panda_lift_demo_to_lerobot_umiThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.1",
"robot_type": null,
"total_episodes": 3,
"total_frames": 1384,
"total_tasks": 3,
"total_videos": 0,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 30,
"splits": {
"train": "0:3"
},
"data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet",
"video_path": null,
"features": {… See the full description on the dataset page: https://huggingface.co/datasets/Project-Rosey/robosuite_panda_lift_demo_to_lerobot_umi.umi_sim_green_cube_v4This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.1",
"robot_type": "xarm6",
"total_episodes": 100,
"total_frames": 18725,
"total_tasks": 1,
"total_videos": 0,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 30,
"splits": {
"train": "0:100"
},
"data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/gellert5225/umi_sim_green_cube_v4.umi-pr17-use-images-false-twocam-1280x720-20260708-183203This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": null,
"total_episodes": 1,
"total_frames": 19,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:1"
},
"data_path": "data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet",
"video_path": null,
"features":… See the full description on the dataset page: https://huggingface.co/datasets/brandonyang/umi-pr17-use-images-false-twocam-1280x720-20260708-183203.umi_sim_green_cube_v6This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.1",
"robot_type": "unknown",
"total_episodes": 150,
"total_frames": 29078,
"total_tasks": 1,
"total_videos": 0,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 30,
"splits": {
"train": "0:150"
},
"data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/gellert5225/umi_sim_green_cube_v6.umi_sim_pickThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.1",
"robot_type": "xarm6",
"total_episodes": 49,
"total_frames": 11531,
"total_tasks": 1,
"total_videos": 0,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 60,
"splits": {
"train": "0:49"
},
"data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/gellert5225/umi_sim_pick.umi_sim_sortThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.1",
"robot_type": "xarm6",
"total_episodes": 48,
"total_frames": 18346,
"total_tasks": 1,
"total_videos": 0,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 60,
"splits": {
"train": "0:48"
},
"data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/gellert5225/umi_sim_sort.umi-pr17-use-images-false-twocam-800x600-20260708-183203This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": null,
"total_episodes": 1,
"total_frames": 33,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:1"
},
"data_path": "data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet",
"video_path": null,
"features":… See the full description on the dataset page: https://huggingface.co/datasets/brandonyang/umi-pr17-use-images-false-twocam-800x600-20260708-183203.umi-pr17-use-images-false-twocam-640x480-20260708-183203This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": null,
"total_episodes": 1,
"total_frames": 51,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:1"
},
"data_path": "data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet",
"video_path": null,
"features":… See the full description on the dataset page: https://huggingface.co/datasets/brandonyang/umi-pr17-use-images-false-twocam-640x480-20260708-183203.stack_cubes_umi_1_franka