Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Emotion Recognition from Facial Expressions Considering Individual Differences in Emotional Intelligence.

Biomimetics (Basel, Switzerland)·2026

Same author

Designing Light for Emotion: A Neurophysiological Approach to Modeling Affective Responses to the Interplay of Color and Illuminance.

Biomimetics (Basel, Switzerland)·2025

Same author

Graph-Driven Micro-Expression Rendering with Emotionally Diverse Expressions for Lifelike Digital Humans.

Biomimetics (Basel, Switzerland)·2025

Same author

Development of an Interactive Digital Human with Context-Sensitive Facial Expressions.

Sensors (Basel, Switzerland)·2025

Same author

Examining the Impact of Digital Human Gaze Expressions on Engagement Induction.

Biomimetics (Basel, Switzerland)·2023

Same author

Recognition of Empathy from Synchronization between Brain Activity and Eye Movement.

Sensors (Basel, Switzerland)·2023

Search research articles

Related Experiment Video

Updated: Sep 16, 2025

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

Emotion Recognition from rPPG via Physiologically Inspired Temporal Encoding and Attention-Based Curriculum Learning.

Changmin Lee¹, Hyunwoo Lee², Mincheol Whang¹

¹Department of Human-Centered Artificial Intelligence, Sangmyung University, Seoul 03016, Republic of Korea.

Sensors (Basel, Switzerland)

|July 12, 2025

Summary

This study introduces a deep learning framework for non-contact emotion recognition using remote photoplethysmography (rPPG). The model shows promise for arousal detection but highlights limitations for valence recognition, suggesting the need for multimodal approaches.

Keywords:

affective computing autonomic nervous system curriculum learning emotion recognition physiological computing remote photoplethysmography sparse attention temporal dynamics

More Related Videos

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception

Published on: August 9, 2024

Assessment of Audio-Tactile Sensory Substitution Training in Participants with Profound Deafness Using the Event-Related Potential Technique

Assessment of Audio-Tactile Sensory Substitution Training in Participants with Profound Deafness Using the Event-Related Potential Technique

Published on: September 7, 2022

Related Experiment Videos

Last Updated: Sep 16, 2025

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception

Published on: August 9, 2024

Assessment of Audio-Tactile Sensory Substitution Training in Participants with Profound Deafness Using the Event-Related Potential Technique

Assessment of Audio-Tactile Sensory Substitution Training in Participants with Profound Deafness Using the Event-Related Potential Technique

Published on: September 7, 2022

Area of Science:

Affective computing
Physiological signal processing
Machine learning for emotion recognition

Background:

Remote photoplethysmography (rPPG) offers non-contact emotion recognition.
Challenges include sparse emotional responses, noise, weak labels, and subtle valence correlates.

Purpose of the Study:

To develop a deep learning framework for robust emotion recognition from rPPG signals.
To address challenges of temporal sparsity, noise, and weak labels in rPPG-based emotion analysis.

Main Methods:

A novel deep learning framework with a Multi-scale Temporal Dynamics Encoder (MTDE).
Adaptive sparse α-Entmax attention and Gated Temporal Pooling for feature aggregation.
A three-phase curriculum learning strategy to manage data complexities.

Main Results:

The temporal-only model achieved competitive arousal recognition accuracy (66.04%) and F1-score (61.97%).
Valence recognition accuracy was lower (62.26%), indicating limitations of unimodal temporal analysis.
Established benchmarks for temporal-only rPPG emotion recognition.

Conclusions:

The proposed framework demonstrates potential for arousal recognition from rPPG.
Unimodal temporal cardiovascular analysis has inherent limitations for valence recognition.
Future research should integrate spatial or multimodal data for comprehensive emotion recognition.