Related Experiment Videos
STFMF-Net: A Hybrid Attention-Driven Multi-View Fusion Framework for Non-Contact Anxiety Recognition Via Facial
IEEE Journal of Biomedical and Health Informatics
|August 7, 2026
Summary
This study introduces a new multi-view method for recognizing anxiety from facial videos, improving accuracy by fusing various physiological signals. The advanced framework offers a precise, non-contact solution for mental health screening.
Area of Science:
- Psychology
- Computer Science
- Biomedical Engineering
Background:
- Non-contact anxiety recognition using facial videos is crucial for large-scale mental health screening.
- Existing methods often rely on limited data like facial expressions or remote photoplethysmography (rPPG), failing to capture anxiety's multi-dimensional nature.
- Anxiety manifests through subtle cues including eye movements, pupil diameter, blinking, and rPPG signals, necessitating a comprehensive approach.
Purpose of the Study:
- To develop a novel multi-view framework for enhanced anxiety recognition from facial videos.
- To address the limitations of single-view approaches by integrating diverse behavioral and physiological data.
- To improve the accuracy and efficiency of non-contact anxiety detection for mental health assessment.
Main Methods:
- A multi-fusion attention net (MFA-Net) was developed for accurate pupil diameter estimation in challenging video conditions.
- A multi-view fusion network (STFMF-Net) with a Hybrid Fusion Module was designed to integrate spatiotemporal and time-frequency features.
- The framework fused data from eye movement trajectories, pupil diameter, eye-blink, head movement, and rPPG signals.
Main Results:
- The proposed multi-view approach achieved a high accuracy of 97.39% and an F1 score of 97.19% on the UBFC-Phys dataset.
- Performance surpassed state-of-the-art methods by 3.06% in accuracy and 2.80% in F1 score.
- Demonstrated the superior effectiveness of multi-view fusion for precise, non-contact anxiety recognition.
Conclusions:
- Multi-view fusion significantly enhances the accuracy of non-contact anxiety recognition from facial videos.
- The developed framework provides a robust technical solution for early anxiety assessment and mental health screening.
- This approach highlights the potential of integrating multiple physiological and behavioral cues for comprehensive mental state monitoring.
Related Concept Videos
01:21Association Areas of the Cortex
Association areas are regions of the cerebral cortex that do not have a specific sensory or motor function. Instead, they integrate and interpret information from various sources to enable higher cognitive processes such as memory, learning, and decision-making. Some key association areas include the following:
Prefrontal Association Area: This area is located in the frontal lobe and is involved in planning, decision-making, and moderating social behavior. It connects with primary motor areas,...
Prefrontal Association Area: This area is located in the frontal lobe and is involved in planning, decision-making, and moderating social behavior. It connects with primary motor areas,...
01:24Facial Feedback Hypothesis
Charles Darwin proposed that facial expressions are an evolutionary adaptation for communication. He argued that these expressions are not influenced by culture but are universal across species. For example, a snarling expression with exposed teeth signals a threat in many animals, including humans. Darwin also suggested that displaying an emotion can intensify the feeling. Smiling, for example, could enhance one's sense of happiness. This idea laid the foundation for understanding the role of...