Leveraging ResNet and label distribution in advanced intelligent systems for facial expression recognition
With the development of AI (Artificial Intelligence), facial expression recognition (FER) is a hot topic in computer vision tasks. Many existing works employ a single label for FER. Therefore, the label distribution problem has not been considered for FER. In addition, some discriminative features c...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
AIMS Press
2023-04-01
|
| Series: | Mathematical Biosciences and Engineering |
| Subjects: | |
| Online Access: | https://www.aimspress.com/article/doi/10.3934/mbe.2023491?viewType=HTML |
| _version_ | 1797822205660233728 |
|---|---|
| author | Zhenggeng Qu Danying Niu |
| author_facet | Zhenggeng Qu Danying Niu |
| author_sort | Zhenggeng Qu |
| collection | DOAJ |
| description | With the development of AI (Artificial Intelligence), facial expression recognition (FER) is a hot topic in computer vision tasks. Many existing works employ a single label for FER. Therefore, the label distribution problem has not been considered for FER. In addition, some discriminative features can not be captured well. To overcome these problems, we propose a novel framework, ResFace, for FER. It has the following modules: 1) a local feature extraction module in which ResNet-18 and ResNet-50 are used to extract the local features for the following feature aggregation; 2) a channel feature aggregation module, in which a channel-spatial feature aggregation method is adopted to learn the high-level features for FER; 3) a compact feature aggregation module, in which several convolutional operations are used to learn the label distributions to interact with the softmax layer. Extensive experiments conducted on the FER+ and Real-world Affective Faces databases demonstrate that the proposed approach obtains comparable performances: 89.87% and 88.38%, respectively. |
| first_indexed | 2024-03-13T10:04:31Z |
| format | Article |
| id | doaj.art-979183e3d51842b2975c14cac8bd1f1d |
| institution | Directory Open Access Journal |
| issn | 1551-0018 |
| language | English |
| last_indexed | 2024-03-13T10:04:31Z |
| publishDate | 2023-04-01 |
| publisher | AIMS Press |
| record_format | Article |
| series | Mathematical Biosciences and Engineering |
| spelling | doaj.art-979183e3d51842b2975c14cac8bd1f1d2023-05-23T01:12:22ZengAIMS PressMathematical Biosciences and Engineering1551-00182023-04-01206111011111510.3934/mbe.2023491Leveraging ResNet and label distribution in advanced intelligent systems for facial expression recognitionZhenggeng Qu 0Danying Niu11. College of Mathematics and Computer Application, Shangluo University, Shaanxi 726000, China 2. Engineering Research Center of Qinling Health Welfare Big Data, Shaanxi 726000, China3. Shangluo Central Hospital, Shaanxi 726000, ChinaWith the development of AI (Artificial Intelligence), facial expression recognition (FER) is a hot topic in computer vision tasks. Many existing works employ a single label for FER. Therefore, the label distribution problem has not been considered for FER. In addition, some discriminative features can not be captured well. To overcome these problems, we propose a novel framework, ResFace, for FER. It has the following modules: 1) a local feature extraction module in which ResNet-18 and ResNet-50 are used to extract the local features for the following feature aggregation; 2) a channel feature aggregation module, in which a channel-spatial feature aggregation method is adopted to learn the high-level features for FER; 3) a compact feature aggregation module, in which several convolutional operations are used to learn the label distributions to interact with the softmax layer. Extensive experiments conducted on the FER+ and Real-world Affective Faces databases demonstrate that the proposed approach obtains comparable performances: 89.87% and 88.38%, respectively.https://www.aimspress.com/article/doi/10.3934/mbe.2023491?viewType=HTMLaffective computingfacial expression recognitiondeep learning |
| spellingShingle | Zhenggeng Qu Danying Niu Leveraging ResNet and label distribution in advanced intelligent systems for facial expression recognition Mathematical Biosciences and Engineering affective computing facial expression recognition deep learning |
| title | Leveraging ResNet and label distribution in advanced intelligent systems for facial expression recognition |
| title_full | Leveraging ResNet and label distribution in advanced intelligent systems for facial expression recognition |
| title_fullStr | Leveraging ResNet and label distribution in advanced intelligent systems for facial expression recognition |
| title_full_unstemmed | Leveraging ResNet and label distribution in advanced intelligent systems for facial expression recognition |
| title_short | Leveraging ResNet and label distribution in advanced intelligent systems for facial expression recognition |
| title_sort | leveraging resnet and label distribution in advanced intelligent systems for facial expression recognition |
| topic | affective computing facial expression recognition deep learning |
| url | https://www.aimspress.com/article/doi/10.3934/mbe.2023491?viewType=HTML |
| work_keys_str_mv | AT zhenggengqu leveragingresnetandlabeldistributioninadvancedintelligentsystemsforfacialexpressionrecognition AT danyingniu leveragingresnetandlabeldistributioninadvancedintelligentsystemsforfacialexpressionrecognition |