Dataset profile
Ego4D
Ego4D is a massive first-person video dataset collected across daily-life scenarios. While not healthcare-specific, it is relevant for embodied AI teams working on egocentric perception, narration, hand-object interaction, and long-form activity understanding.
Overview
Ego4D is a massive first-person video dataset collected across daily-life scenarios. While not healthcare-specific, it is relevant for embodied AI teams working on egocentric perception, narration, hand-object interaction, and long-form activity understanding.
What's included
- Long-form first-person video
- Narrations and benchmark annotations
- Selected audio, gaze, stereo, and 3D environment data
Tasks / activities
Modalities
Collection methodology
Camera wearers captured daily activities in natural environments across multiple countries and locations.
Potential applications
- Egocentric representation learning
- Hand-object understanding
- Narration and temporal localization
- First-person pretraining
Access & licensing
This is a third-party public/research dataset. We do not own or distribute this dataset. Access and usage are governed by the original publisher's terms.
Access requires the original Ego4D license process; usage follows publisher terms.
Limitations / opportunities for custom data
Custom collection can focus on healthcare roles, care facilities, controlled task protocols, and modalities absent from broad daily-life video.
Need data beyond this benchmark?