README.md

---
pretty_name: "EgoSuite-Open100K - EgoDemo"
language:
- en
license: other
license_name: commercial-training-no-resale-v1.0
size_categories:
- 1K<n<10K
task_categories:
- video-classification
tags:
- video
- egocentric-video
- embodied-ai
- human-demonstration
- public-demo
- provenance
- human-pose
- hand-pose
- body-pose
- wrist-camera
- multimodal
- lerobot
- mcap
- raw-video
- robotics
gated: true
viewer: false
extra_gated_heading: Get access to EgoDemo
extra_gated_description: >-
  We'd love to know a little about who's using the data.
  Everyone is welcome, whether you're at a lab, a company, or working on your own.
extra_gated_fields:
  "Name": text
  "Affiliation (optional)": text
  "What are you planning to use it for?": text
extra_gated_button_content: Get access
---

<div align="center">
<picture>
<source media="(prefers-color-scheme: dark)" srcset="https://huggingface.co/spaces/LightwheelAI/README/resolve/main/assets/Lightwheel_white.png" />
<source media="(prefers-color-scheme: light)" srcset="https://huggingface.co/spaces/LightwheelAI/README/resolve/main/assets/Lightwheel.png" />
<img src="https://huggingface.co/spaces/LightwheelAI/README/resolve/main/assets/Lightwheel.png" alt="Lightwheel" width="300" />
</picture>
</div>

<h1 align="center">EgoDemo</h1>
<p align="center">A 50-hour sample from EgoSuite-Open100K, covering every annotated subset plus two raw-video variants.</p>
<p align="center">
  <a href="https://huggingface.co/collections/LightwheelAI/egosuite-open100k">Collection</a> ·
  <a href="https://huggingface.co/datasets/LightwheelAI/EgoStandard">EgoStandard</a> ·
  <a href="https://huggingface.co/datasets/LightwheelAI/EgoPro">EgoPro</a> ·
  <a href="https://egosuite100k.lightwheel.ai/">Project page</a>
</p>

[![Explore the EgoSuite-Open100K dataset](assets/egosuite-dataset-viewer.jpg)](https://egosuite100k.lightwheel.ai/)

<p align="center"><a href="https://egosuite100k.lightwheel.ai/"><strong>Open the interactive dataset explorer ↗</strong></a></p>

## Overview

EgoDemo is the developer-friendly entry point to EgoSuite-Open100K. Unlike EgoStandard and EgoPro, its data is stored directly in this dataset repository. The annotated samples are available in both LeRobot v3 and MCAP; the raw-video samples are MP4.

| Variant | Camera views | Pose annotations | Formats |
|---|---|---|---|
| <code>EgoStand</code> | Head | Hands | LeRobot, MCAP |
| <code>EgoStand-body</code> | Head | Hands, full body | LeRobot, MCAP |
| <code>EgoProStandard</code> | Head, wrist | Hands | LeRobot, MCAP |
| <code>EgoProStandard-body</code> | Head, wrist | Hands, full body | LeRobot, MCAP |
| <code>EgoRaw</code> | Head | None | MP4 |
| <code>EgoProRaw</code> | Head, wrist | None | MP4 |

The 15,000+ task and scene figures associated with EgoSuite-Open100K describe the full collection, not this 50-hour sample.

## Download

Accept the access terms on this page, then authenticate with the Hugging Face CLI:

    pip install -U huggingface_hub
    hf auth login

Download the complete repository:

    hf download LightwheelAI/EgoDemo \
      --repo-type dataset \
      --local-dir ./EgoDemo

Use <code>--include</code> to download only the data you need:

    # One annotated LeRobot variant
    hf download LightwheelAI/EgoDemo \
      --repo-type dataset \
      --include "EgoStand/lerobot/**" \
      --local-dir ./EgoDemo

    # Raw head-view video
    hf download LightwheelAI/EgoDemo \
      --repo-type dataset \
      --include "EgoRaw/mp4/**" \
      --local-dir ./EgoDemo

See the [Hugging Face download documentation](https://huggingface.co/docs/huggingface_hub/en/guides/download) for caching, filtering, and dry-run options.

## Repository Layout

    EgoDemo/
    ├── README.md
    ├── EgoStand/
    │   ├── README.md
    │   ├── lerobot/
    │   ├── mcap/
    │   └── manifests/
    ├── EgoStand-body/
    │   ├── README.md
    │   ├── lerobot/
    │   ├── mcap/
    │   └── manifests/
    ├── EgoProStandard/
    │   ├── README.md
    │   ├── lerobot/
    │   ├── mcap/
    │   └── manifests/
    ├── EgoProStandard-body/
    │   ├── README.md
    │   ├── lerobot/
    │   ├── mcap/
    │   └── manifests/
    ├── EgoRaw/
    │   ├── README.md
    │   ├── mp4/
    │   └── manifests/
    └── EgoProRaw/
        ├── README.md
        ├── mp4/
        └── manifests/

An absent format directory means that format is not provided for the variant. Variant-specific details are documented in the README inside each directory.

## Data Contract

LeRobot and MCAP are alternative representations of the same annotated episodes. The shared <code>source_episode_uuid</code> connects formats; do not add their durations or episode counts together.

Pose data uses <code>float32</code> world coordinates:

| Value | Shape |
|---|---:|
| Left/right hand positions | <code>(21, 3)</code> |
| Left/right hand rotations | <code>(21, 4)</code> |
| Full-body positions | <code>(22, 3)</code> |
| Full-body rotations | <code>(22, 4)</code> |

Quaternions use <code>(qw, qx, qy, qz)</code> ordering. Temporal event-level semantic annotations are included for the four annotated variants; the raw-video variants contain no pose or semantic annotations.

Manifest rows provide stable identifiers, source provenance, task and scene metadata, duration, repository-relative paths, modality flags, format information, and SHA-256 checksums. Raw-video manifest rows omit source Sub-SKU and pose fields.

## Format Versions

| Item | Version |
|---|---|
| Dataset release | <code>v1.0.0</code> |
| LeRobot | <code>v3.0</code> |
| MCAP container | <code>MCAP0</code> |
| MCAP schema encoding | Embedded Protobuf |
| Raw video | MP4 + JSON |

## Privacy and Use

All participants provided informed consent covering collection, annotation, controlled release, and uses permitted by the dataset license. Recordings were de-identified before release. Use of the data is governed by the repository license and access terms.

For dataset questions, open a discussion in this repository. For other inquiries, contact [contact@lightwheel.ai](mailto:contact@lightwheel.ai).