Related Experiment Video
Updated: Oct 15, 2025
08:05
Design and Analysis for Fall Detection System Simplification
Published on: April 6, 2020
10.9K
A high-fidelity residential building occupancy detection dataset
Margarite Jacoby1, Sin Yong Tan2, Gregor Henze3,4,5
1University of Colorado Boulder, Department of Civil, Environmental and Architectural Engineering, Boulder, 80309-0428, United States. margarite.jacoby@colorado.edu.
Scientific Data
|October 29, 2021
Summary
A new dataset captures home occupancy using images, audio, and environmental data. This resource supports research in energy efficiency and indoor environmental quality.
Area of Science:
- Building Science
- Data Science
- Human-Computer Interaction
Background:
- Occupancy detection is crucial for energy efficiency and indoor environmental quality.
- Limited publicly available datasets exist for comprehensive occupancy monitoring in homes.
Purpose of the Study:
- To develop and deploy a data acquisition system for capturing multi-modal occupancy data in single-family residences.
- To create a publicly available dataset for research applications.
Main Methods:
- Developed the mobile human presence detection (HPDmobile) system.
- Deployed the system in six homes for at least one month each, collecting data from multiple locations concurrently.
- Captured grayscale images, processed audio, indoor environmental readings, and ground truth occupancy status.
Main Results:
- Generated a comprehensive dataset of occupancy-related modalities.
- Ensured privacy by processing images and audio data.
- Dataset includes high-frequency image and lower-frequency environmental data.
Conclusions:
- The HPDmobile dataset provides a valuable resource for advancing research in energy efficiency and indoor environmental quality.
- The developed system and dataset facilitate the study of human presence in residential settings.
