August 25, 2025SensorsOpen Access

Development of Machine-Learning-Based Facial Thermal Image Analysis for Dynamic Emotion Sensing

View Full Paper

Ask AI

Bookmark

Authors

BTBudu Tang WSWataru SatoKyoto University YKYasutomo Kawanishi

Discussion

Member takes

Implication

Study demonstrates non-linear emotion sensing using facial thermal images in 20 participants, suggesting enhanced ML techniques may improve mental health assessments.

Key Points

Machine learning outperformed traditional linear regression in estimating emotional arousal from facial thermal images.
ML models like ResNet-34 showed significant accuracy in predicting emotional responses from specific facial areas.
Analysis included saliency maps for interpreting complex relationships between temperature changes and emotions.
Findings highlight the potential of non-invasive emotion sensing in applications like mental health and human-computer interaction.

Cite This Study

Tang et al. (2025) studied this question.

synapsesocial.com/papers/68af7c8d7567bf4f94ff403c https://doi.org/10.3390/s25175276

View Full Paper

Ask AI

Bookmark

Implication

Study demonstrates non-linear emotion sensing using facial thermal images in 20 participants, suggesting enhanced ML techniques may improve mental health assessments.

Development of Machine-Learning-Based Facial Thermal Image Analysis for Dynamic Emotion Sensing

Discussion

Member takes

Implication

Key Points

Cite This Study

Development of Machine-Learning-Based Facial Thermal Image Analysis for Dynamic Emotion Sensing

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider