HACER: An Integrated Remote Monitoring Platform for the Elderly

Document Type

Conference Proceeding

Publication Title

2023 International Conference on Intelligent Metaverse Technologies and Applications, iMETA 2023

Abstract

Recognizing human actions and emotions using video analysis has great potential for improving the quality of life for the elderly. However, current datasets used for the recognition typically focus on either emotion recognition or action recognition, limiting the scope of research investigating the interdependence between actions and emotions. To address this limitation, we present an end-to-end process for jointly performing human action and emotion recognition by simultaneously extracting action-specific and emotion-specific features from video input. Our proposed approach aims to develop and test machine learning models for recognizing human actions and emotions in smart environments for the elderly to monitor remotly. Additionally, we propose the Human ACtion and Emotion Recognition dataset, HACER, a unique dataset that jointly incorporates both emotion and action labels, providing valuable insights into the interplay between emotions and actions. Our proposed dataset fills a crucial gap in the existing literature, enabling the development of machine learning models that can recognize and classify both action and emotion categories at once, advancing the field of multimodal recognition with only one sensor. The dataset is publicly available at: https://www.kaggle.com/datasets/siwarammar/hacer-human-action-and-emotion-recognition

DOI

10.1109/iMETA59369.2023.10294645

Publication Date

10-2023

Keywords

Emotion recognition, Limiting, Machine learning, Medical services, Feature extraction, Data models, Older adults

Comments

IR conditions: non-described

Link to Full Text

Share

COinS