Related Experiment Video
Updated: Jun 26, 2026
06:25
Capturing Representative Hand Use at Home Using Egocentric Video in Individuals with Upper Limb Impairment
Published on: December 23, 2020
The use of computer vision techniques to augment home based sensorised environments.
Zdenka Uhríková1, Chris D Nugent, Václav Hlavác
1Faculty of Electrical Engineering, Czech Technical University in Prague, Prague 2, Czech Republic. uhrikz1@cmp.felk.cvut.cz
Summary
Computer vision enhances smart home technology by improving task identification in multi-person environments. This approach addresses challenges in sensor data and facilitates independent living.
Area of Science:
- Computer Science
- Artificial Intelligence
- Human-Computer Interaction
Background:
- Home-based technology supports independent living.
- Challenges include distinguishing tasks among multiple occupants and managing sensor data uncertainty.
Purpose of the Study:
- To explore computer vision techniques for behavior assessment in sensorized home environments.
- To evaluate vision processing's effectiveness with sensor data for multi-occupancy scenarios.
Main Methods:
- Utilized computer vision techniques, specifically color-based tracking algorithms.
- Integrated video data with existing sensor-based data.
- Assessed performance in detecting and recognizing multiple individuals and their tasks.
Main Results:
- Computer vision integration significantly improved task identification accuracy.
- Color-based tracking successfully detected and recognized multiple people.
- Effectively handled multi-occupancy situations.
Conclusions:
- Computer vision offers a valuable tool for enhancing smart home behavior assessment.
- Combining vision and sensor data overcomes key challenges in multi-person environments.
- This technology advances the potential for supported independent living.
