Emotions Recognition Using Multimodal Spontaneous Emotion Database and Deep Learning Technology
Bibliographic record
Abstract
Facial expression Recognition (FER) has growing significance in diverse fields such as psychology, medicine, sports, and entertainment. FER is used in the medical field to recognize signs of depression, anxiety, and autism. FER also finds its niche in self-driving cars to observe signs of fatigue and distress in a driver and provide timely intervention to enhance transport safety. Facial expressions combined with other modalities offer great insight into the emotional state and its triggers. Computer vision, machine learning, and deep learning methods have recently gained widespread attention in detecting and classifying spontaneous facial expressions. Static images and video sequences in 2D have extensively been used for FER and emotion recognition. However, only a few algorithms combine 2D video sequences and multimodal data to detect and classify emotions. To this end, this research aims to develop a deep-learning model for classifying emotions using Karolinska Directed Emotional Face (KDEF) and the Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS). A CNN-RNN model is built to classify facial expressions for synthetic video sequence data generated using the KDEF dataset. This model is extended to features extracted from the RAVDESS video dataset. Furthermore, A Transformer model with a dual-head self-attention layer is created to identify the frames with the most useful information for classification. Finally, a late fusion architecture is used to merge the posteriors of the static audio, static video, and Transformer models to create a multimodal classification model.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".