Video · 3D / Point cloud / Manufacturing & Industrial
Egocentric Video Dataset — Pico + Motion Trackers Setup (2,321 hours)
Advertised by providerLicensing rights not independently verifiedPaid licence
Overview
Egocentric Video Dataset — Pico + Motion Trackers Setup (2,321 hours) is a video and 3d / point cloud dataset. The provider states a volume of 2,321 hours. Typical applications include object detection and robotics & embodied AI. Relevant sectors: Manufacturing & Industrial. The listing mentions keypoints. It is offered as a paid dataset; pricing details are provided on request.
Primary applications
What the listing offers
- Stated volume of 2,321 hours.
- Listing mentions keypoints.
Overview This egocentric dataset is a commercial collection of 2,321 hours of first-person videos produced by the provider for Physical AI, robotic systems, manipulation tasks, and egocentric vision research. All recordings capture human motions during daily household activities in real home environments, paired with motion capture data from body-worn trackers and quaternion-based orientation from onboard sensor fusion. This combination of egocentric video and synchronized motion tracking supports 3D pose estimations and egocentric tracking across first-person perspectives, giving robotic and computer vision teams training data grounded in genuine, real-world human activity rather than synthetic footage. What's Included - 2,321 hours of first-person videos captured with a Pico 4 Ultra VR headset - Synchronized body-worn motion trackers delivering full-body human motions - Quaternion orientation from onboard sensor-fusion for precise 3D pose reconstruction - IMU signals — accelerometer, gyroscope, and magnetometer streams - A .txt labeling file per recording with data annotations Hardware & Recording Conditions The Pico + Motion Trackers setup focuses purely on the egocentric viewpoint, capturing how hands motions unfold from the wearer's own perspective. Recordings span three distinct conditions: 1. Natural speed — hands appear in frame only when needed, reflecting unscripted daily activity 2. Slow motion — both hands always in frame, capturing detailed kinematics for fine-grained
Related datasets
More in Video & Physical AI →Video · 3D / Point cloudAdvertised
Egocentric Video Dataset — 4,050 Hours of First-Person Videos for Physical AI, Robotics & Egocentric Vision
Overview The Egocentric Video Dataset is a large-scale, multimodal collection produced by the provider, containing 4,050 hours of first-person egocentric videos of daily…
- Language
- —
- Volume
- 4,050 h
- Domain
- Manufacturing & Industrial
- Use case
- Object detection
USD 30,000
More video & physical ai datasets
All 302 video & physical ai datasets →- Egocentric Video: Event & Activity Venue Setup, Teardown & Operations | Synchronized Stereo Camera + 400Hz IMU | Physical AI, Embodied data
- Egocentric video of household chores, annotated and labeled
- Egocentric video of indoor and outdoor human activities
- Egocentric videos with intentional errors, detection-correction paired
- Egocentric Warehousing & Distribution Video Data: Picking, Packing & Receiving | Hand-Keypoint & SLAM/VIO Tracking | Robot Learning, Physical AI
- Embodied AI Data Solutions | Data Collection & Annotation | AI Robot | Multimodal Data | Video Data
- English animation spanning history, science, nature, culture
- English animation video across diverse genres