datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
pusht_keypointsThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "unknown",
"total_episodes": 206,
"total_frames": 25650,
"total_tasks": 1,
"chunks_size": 1000,
"fps": 10,
"splits": {
"train": "0:206"
},
"data_path": "data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet",
"video_path": null,
"features": {
"observation.state": {
"dtype":… See the full description on the dataset page: https://huggingface.co/datasets/lerobot/pusht_keypoints.hand-keypoints
Dataset Card for Image Hand Keypoint Detection
This is a FiftyOne dataset with 846 samples.
Note: The images here are from the test set of the original dataset and parsed into FiftyOne format.
Installation
If you haven't already, install FiftyOne:
pip install -U fiftyone
Usage
import fiftyone as fo
from fiftyone.utils.huggingface import load_from_hub
# Load the dataset
# Note: other available arguments include 'max_samples', etc
dataset =… See the full description on the dataset page: https://huggingface.co/datasets/Voxel51/hand-keypoints.mouse_keypoints
YOLO Pose Dataset (ADPT-derived)
This dataset contains images and YOLO-format pose annotations for training Ultralytics YOLO pose models. The original images/labels are derived from the ADPT dataset (see source below). This release provides a YOLO-ready layout plus a simple data.yaml for immediate training and evaluation.
Source / Attribution
Original dataset: ADPT (Tang Guoling et al.)
Upstream repository: https://github.com/tangguoling/ADPT
This Hugging Face dataset is… See the full description on the dataset page: https://huggingface.co/datasets/healthonrails/mouse_keypoints.pusht_keypointsThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.0",
"robot_type": "2d pointer",
"total_episodes": 206,
"total_frames": 25650,
"total_tasks": 1,
"total_videos": 0,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 10,
"splits": {
"train": "0:206"
},
"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/lilkm/pusht_keypoints.rugby-pitch-keypoints-detection-v3pusht-keypoints-only-diffThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.1",
"robot_type": "unknown",
"total_episodes": 206,
"total_frames": 25650,
"total_tasks": 1,
"total_videos": 0,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 10,
"splits": {
"train": "0:206"
},
"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/the-future-dev/pusht-keypoints-only-diff.gugugaga-keypoints-v1
gugugaga-keypoints-v1
Synthetic 18-point 2D keypoint dataset rendered from a chibi character (gugugaga) for pose / keypoint model training.
Summary
Item
Value
Samples
450
Poses
25 anchors
Cameras
front, back, side_r, side_l, top, bottom
Pitch
eye, high, low
Image size
720 × 1280
Schema
gugugaga_kp18_v1 (18 keypoints)
Grid: 25 pose × 6 camera × 3 pitch = 450.
Layout
images/{pose_id}/{stem}_rgb.png… See the full description on the dataset page: https://huggingface.co/datasets/todo1111/gugugaga-keypoints-v1.pusht-keypoints-expanded-diffThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.1",
"robot_type": "unknown",
"total_episodes": 206,
"total_frames": 25650,
"total_tasks": 1,
"total_videos": 0,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 10,
"splits": {
"train": "0:206"
},
"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/the-future-dev/pusht-keypoints-expanded-diff.SoccerNet_Field_Keypointstennis_court_keypoints_datasetThis is just a re-upload of this dataset: https://github.com/yastrebksv/TennisCourtDetector that was originally uploaded to google drive. I uploaded it here for easier access.
The dataset consists of 8841 images, which were separeted to train set (75%) and validation set (25%). Each image has 14 annotated points. The resolution of images is 1280×720. This dataset contains all court types (hard, clay, grass).
contact me for removal gholamrezadar@gmail.com
COCO_keypointspusht_v2_keypointsThis dataset was created using LeRobot.
Info
meta/info.json
{
"codebase_version": "v2.0",
"robot_type": "2d pointer",
"total_episodes": 206,
"total_frames": 25650,
"total_tasks":1,
"total_videos": 0,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 10,
"splits": {
"train": "0:206"
},
"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/cadene/pusht_v2_keypoints.processed-episodes-gestures-keypoints3dT2I-Keypoints-Eval
T2I-Keypoints-Eval Dataset
A Bilingual Text-to-Image Keypoints Evaluation Benchmark
Linqing Wang ·
Ximing Xing ·
Yiji Cheng ·
Zhiyuan Zhao ·
Jiale Tao ·
QiXun Wang ·
Ruihuang Li ·
Comi Chen ·
Xin Li ·
Mingrui Wu ·
Xinchi Deng ·
Chunyu Wang† ·
Qinglin Lu*
Tencent Hunyuan
†Project Lead · *Corresponding Author
Overview
T2I-Keypoints-Eval is a comprehensive bilingual evaluation dataset designed to assess text-to-image models' ability to generate images… See the full description on the dataset page: https://huggingface.co/datasets/PromptEnhancer/T2I-Keypoints-Eval.hand-keypoints-dataset
Dataset Card for Hand Keypoint Detection
Dataset Description
This dataset contains hand images annotated with keypoints for pose estimation.
Data Format
Images are in RGB format
Annotations include bounding boxes and keypoint coordinates
Keypoints include visibility flags
Keypointscitysample-vehicle-keypoints-24pt
City Sample Vehicle Keypoints (24-point, synthetic)
A synthetic vehicle-keypoint dataset rendered inside Epic's City Sample
("Matrix Awakens") with Unreal Engine 5.6 and Movie Render Queue. Vehicles are
placed on the real ZoneGraph road network, and every visible vehicle is labelled
with a 24-point anatomical keypoint schema plus a mesh-bounds bounding box.
Generated by the open pipeline at
kiselyovd/ue5-vehicle-synth
(full method, code, and engineering write-up in the repo… See the full description on the dataset page: https://huggingface.co/datasets/kiselyovd/citysample-vehicle-keypoints-24pt.footwork-detection-keypoints
Footwork Detection Keypoints Dataset
Dataset Description
This dataset was created from scratch for research and development in automated footwork detection and tactical analysis using computer vision and machine learning.
Unlike datasets collected from existing public benchmarks, this dataset was specifically constructed and organized by the authors for the footwork detection task.
Dataset Creation
The dataset was collected, processed, and annotated… See the full description on the dataset page: https://huggingface.co/datasets/Pranathi196/footwork-detection-keypoints.pusht_v2_keypoints
pusht_v2_keypoints (TsFile)
Apache TsFile version of cadene/pusht_v2_keypoints.
Overview
The keypoints variant of the PushT benchmark, formatted with
LeRobot. A 2D pointer agent pushes a
T-shaped block onto a T-shaped target region. Each frame records the agent state,
the action, the environment keypoint state, and the step reward / success flag.
Robot: 2d pointer (2D agent).
Task: "Push the T-shaped blue block onto the T-shaped green target surface." (1 task).… See the full description on the dataset page: https://huggingface.co/datasets/THULab/pusht_v2_keypoints.coco_keypoints
Dataset Card for "COCO Keypoints"
Quick Start
Usage
>>> from datasets.load import load_dataset
>>> dataset = load_dataset('whyen-wang/coco_keypoints')
>>> example = dataset['train'][0]
>>> print(example)
{'image': <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=640x360>,
'bboxes': [
[339.8800048828125, 22.15999984741211,
153.8800048828125, 300.7300109863281],
[471.6400146484375, 172.82000732421875,
35.91999816894531, 48.099998474121094]]… See the full description on the dataset page: https://huggingface.co/datasets/whyen-wang/coco_keypoints.pusht_image_keypointsThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "unknown",
"total_episodes": 206,
"total_frames": 25650,
"total_tasks": 1,
"chunks_size": 1000,
"fps": 10,
"splits": {
"train": "0:206"
},
"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",
"features": {… See the full description on the dataset page: https://huggingface.co/datasets/the-future-dev/pusht_image_keypoints.pusht_keypoints_imagesThis dataset was created using 🤗 LeRobot.
pusht_keypoints_images_suboptimal_trajectoriesThis dataset was created using 🤗 LeRobot.
cyclist-intention-2d-keypoints
Cyclist Behavior 2D Keypoints V5
Version 2.0.0 is a complete replacement of the earlier intersection release. It contains
de-identified cyclist skeleton scene clips for bus-stop/road behavior prediction.
Summary
32 participants, 559 scene clips, 99,734 frames
33 two-dimensional pose landmarks (66 coordinates)
Stage1 labels: head look, upper-limb rotation, and pedaling intervals
Stage2 labels: straight, yield, and overtake
canonical Fold 1 participant-disjoint… See the full description on the dataset page: https://huggingface.co/datasets/Kheyro/cyclist-intention-2d-keypoints.pusht_keypointshow2sign_keypointspusht_keypointsThis dataset was created using LeRobot.
Info
meta/info.json
{
"codebase_version": "v2.0",
"robot_type": "unknown",
"total_episodes": 206,
"total_frames": 25650,
"total_tasks":1,
"total_videos": 0,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 10,
"splits": {
"train": "0:206"
},
"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/iantc104/pusht_keypoints.pusht_keypointsThis dataset was created using LeRobot.
Info
meta/info.json
{
"codebase_version": "v2.0",
"robot_type": "unknown",
"total_episodes": 206,
"total_frames": 25650,
"total_tasks":1,
"total_videos": 0,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 10,
"splits": {
"train": "0:206"
},
"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/brysonjones/pusht_keypoints.tennis_court_keypoints_datasetThis is just a re-upload of this dataset: https://github.com/yastrebksv/TennisCourtDetector that was originally uploaded to google drive. I uploaded it here for easier access.
The dataset consists of 8841 images, which were separeted to train set (75%) and validation set (25%). Each image has 14 annotated points. The resolution of images is 1280×720. This dataset contains all court types (hard, clay, grass).
contact me for removal gholamrezadar@gmail.com
facial-keypoints