Dataset profile

African Actions - Health Domain

African Actions - Health Domain is an egocentric dataset of health-domain activities recorded in Nigeria and paired with fine-grained action-object annotations and multilingual descriptions. It is relevant for embodied AI teams seeking more geographically diverse procedural data.

Source institutionMozilla Data Collective
HealthcareEgocentric

Overview

African Actions - Health Domain is an egocentric dataset of health-domain activities recorded in Nigeria and paired with fine-grained action-object annotations and multilingual descriptions. It is relevant for embodied AI teams seeking more geographically diverse procedural data.

What's included

  • Egocentric MP4 recordings
  • JSON annotations with action, object, tool, and secondary object fields
  • English plus Igbo, Hausa, or Yoruba natural-language descriptions

Tasks / activities

health-domain proceduresfine-grained action-object annotationmultilingual procedural descriptions

Modalities

Egocentric VideoRGBLanguageAnnotations

Collection methodology

Health-domain tasks were recorded from a first-person or near-first-person perspective in Nigerian real-world settings and reviewed through a multi-stage quality process.

Potential applications

  • Geographically diverse healthcare activity modeling
  • Fine-grained action-object recognition
  • Multilingual video-language learning
  • Embodied procedural understanding

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 and use are governed by Mozilla Data Collective and the dataset publisher terms.

Limitations / opportunities for custom data

Teams may need greater scale, additional care roles, different geographies, synchronized sensors, or buyer-specific task protocols.

Need data beyond this benchmark?

We design custom healthcare and caregiving collections around the tasks, experts, environments, and modalities your model requires.

Request Similar Data