Abstract
n emotion recognition, the multimodal feature fu-
sion approach for facial expression recognition is useful due to
its versatility and adaptability. It leads to improved model per-
formance by capturing information from different modalities. In
this study, we employ feature-level fusion, integrating CNN and
HOG features. To predict continuous valence and arousal values,
we utilize a Feedforward neural network and Gradient Boosting.
Performance evaluation is conducted using Mean Squared Error
(MSE) and Root Mean Squared Error (RMSE). The paper
presents experiments using the ADFES dataset, considering low,
medium, and high intensities, as well as an augmented video
dataset. The results shows that instead of relying on complex
models, accuracy can be achieved by combining various types of
features with appropriate hyperparameter settings and tuning.
This approach is not only cost-effective in terms of computation
but also robust and computationally efficient.
sion approach for facial expression recognition is useful due to
its versatility and adaptability. It leads to improved model per-
formance by capturing information from different modalities. In
this study, we employ feature-level fusion, integrating CNN and
HOG features. To predict continuous valence and arousal values,
we utilize a Feedforward neural network and Gradient Boosting.
Performance evaluation is conducted using Mean Squared Error
(MSE) and Root Mean Squared Error (RMSE). The paper
presents experiments using the ADFES dataset, considering low,
medium, and high intensities, as well as an augmented video
dataset. The results shows that instead of relying on complex
models, accuracy can be achieved by combining various types of
features with appropriate hyperparameter settings and tuning.
This approach is not only cost-effective in terms of computation
but also robust and computationally efficient.
Original language | English |
---|---|
Title of host publication | The 25th IEEE International Conference on Industrial Technology |
Publisher | IEEE |
Publication status | Accepted/In press - 25 Mar 2024 |
Event | The 25th IEEE International Conference on Industrial Technology - DoubleTree by Hilton Bristol City Centre, Redcliffe Way, Redcliffe, Bristol, BS1 6NJ, Bristol, United Kingdom Duration: 25 Mar 2024 → 27 Mar 2024 Conference number: 25 https://icit2024.ieee-ies.org/venueAndAccomodation.html |
Conference
Conference | The 25th IEEE International Conference on Industrial Technology |
---|---|
Abbreviated title | ICIT 2024 |
Country/Territory | United Kingdom |
City | Bristol |
Period | 25/03/24 → 27/03/24 |
Internet address |
Keywords
- Emotion recognition
- Feature Fusion
- CNN-HOG
- Valence-Arousal Space
- CNN
- HOG
- Feature-Level-Fusion
- Multimodal Fusion