10000-Hour-Egocentric-Video-Dataset

Description

This dataset contains 10,000 hours of egocentric multimodal data collected from diverse real-world environments, including residential, retail, and office scenarios. It covers a wide range of human activities and manipulation tasks, such as meal preparation, cleaning, storage, garment care, merchandising, and object picking. Each sample includes synchronized 4K stereo video, camera calibration parameters, 76-point full-body pose annotations, and fine-grained step-by-step action sequence labels. The dataset is suitable for robot learning, manipulation policy development, and Vision-Language-Action (VLA) models.

For more details, please refer to the link: https://www.nexdata.ai/datasets/embodied-ai/2145=Huggingface

Data size

10,000-Hour Egocentric Full-Body Multimodal Dataset

Data Distribution

Covers residential, retail & office scenarios (kitchen, bedroom, living room, supermarket, office) with diverse real-life tasks: meal prep, cleaning, storage, garment care, merchandising & picking

Data Content

Each sample includes spatiotemporally aligned 4K stereo video, camera calibration params, 76-point full-body pose & step-by-step annotations

Capture Solution

Adopts PICO 4 Ultra head-mounted stereo camera + wrist & ankle IMU motion capture solution

Data Annotation

Supports dense semantic & action-level annotations; all data passes multi-stage quality control reviews

Data Quality

Supports 4096×1536 / 30fps HD video output, tracks 24 torso joints and 52 hand joints, with frame-wise dense annotations and full-process quality control

Full Dataset Access

The complete dataset is available upon request. Reach out to us to learn more and submit an access request.