Key result
A deep learning architecture combining 1DCNN and LSTM with contactless PPG signals demonstrated excellent performance in binary classification of valence and arousal using 5-second signal segmentation.
Why the study?
Traditional automatic recognition of human emotions relies on sensors placed on the body, but contactless photoplethysmography signal extraction from facial video footage offers an alternative approach.
Population
A widely used emotional database
Comparison
Various techniques for extracting PPG signals combined with a 1DCNN and LSTM architecture
Authors
Loading...
Supports feasibility of contactless PPG emotion classification; leaves open clinical validation and accuracy.
Contactless PPG signal extraction combined with deep learning can effectively classify human emotions based on valence and arousal.
VERMA et al. (2023) studied Human emotions. Contactless PPG signal extraction and deep learning (1DCNN + LSTM) was evaluated on Binary classification of valence and arousal. A deep learning architecture combining 1DCNN and LSTM with contactless PPG signals demonstrated excellent performance in binary classification of valence and arousal using 5-second signal segmentation.
Synapse has enriched 2 closely related papers on similar clinical questions. Consider them for comparative context: