A joint learning method for low-light facial expression recognition
Abstract Existing facial expression recognition (FER) methods are mainly devoted to learning discriminative features from normal-light images. However, their performance drops sharply when they are used for low-light images. In this paper, we propose a novel low-light FER framework (termed LL-FER) t...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2025-01-01
|
| Series: | Complex & Intelligent Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s40747-024-01762-z |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823861513982574592 |
|---|---|
| author | Yuanlun Xie Jie Ou Bihan Wen Zitong Yu Wenhong Tian |
| author_facet | Yuanlun Xie Jie Ou Bihan Wen Zitong Yu Wenhong Tian |
| author_sort | Yuanlun Xie |
| collection | DOAJ |
| description | Abstract Existing facial expression recognition (FER) methods are mainly devoted to learning discriminative features from normal-light images. However, their performance drops sharply when they are used for low-light images. In this paper, we propose a novel low-light FER framework (termed LL-FER) that can simultaneously enhance the images and recognition tasks of low-light facial expression images. Specifically, we first meticulously design a low-light enhancement network (LLENet) to recover expressions images’ rich detail information. Then, we design a joint loss to train the LLENet with FER network in a cascade manner, so that the FER network can guide the LLENet to gradually perceive and restore discriminative features which are useful for FER during the training process. Extensive experiments show that the LLENet not only achieves competitive results both quantitatively and qualitatively, but also in the LL-FER framework, which can produce results more suitable for FER tasks, further improving the performance of the FER methods. |
| format | Article |
| id | doaj-art-e4f3dfc23c234d22b0026c435fd519ec |
| institution | Kabale University |
| issn | 2199-4536 2198-6053 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Springer |
| record_format | Article |
| series | Complex & Intelligent Systems |
| spelling | doaj-art-e4f3dfc23c234d22b0026c435fd519ec2025-02-09T13:00:56ZengSpringerComplex & Intelligent Systems2199-45362198-60532025-01-0111211710.1007/s40747-024-01762-zA joint learning method for low-light facial expression recognitionYuanlun Xie0Jie Ou1Bihan Wen2Zitong Yu3Wenhong Tian4College of Electronic Information and Electrical Engineering, Chengdu UniversitySchool of Information and Software Engineering, University of Electronic Science and Technology of ChinaThe School of Electrical and Electronic Engineering, Nanyang Technological UniversityThe Great Bay UniversitySchool of Information and Software Engineering, University of Electronic Science and Technology of ChinaAbstract Existing facial expression recognition (FER) methods are mainly devoted to learning discriminative features from normal-light images. However, their performance drops sharply when they are used for low-light images. In this paper, we propose a novel low-light FER framework (termed LL-FER) that can simultaneously enhance the images and recognition tasks of low-light facial expression images. Specifically, we first meticulously design a low-light enhancement network (LLENet) to recover expressions images’ rich detail information. Then, we design a joint loss to train the LLENet with FER network in a cascade manner, so that the FER network can guide the LLENet to gradually perceive and restore discriminative features which are useful for FER during the training process. Extensive experiments show that the LLENet not only achieves competitive results both quantitatively and qualitatively, but also in the LL-FER framework, which can produce results more suitable for FER tasks, further improving the performance of the FER methods.https://doi.org/10.1007/s40747-024-01762-zFacial expression recognitionLow-light image enhancementHigh-level vision taskCross high- and low-level learningJoint training |
| spellingShingle | Yuanlun Xie Jie Ou Bihan Wen Zitong Yu Wenhong Tian A joint learning method for low-light facial expression recognition Complex & Intelligent Systems Facial expression recognition Low-light image enhancement High-level vision task Cross high- and low-level learning Joint training |
| title | A joint learning method for low-light facial expression recognition |
| title_full | A joint learning method for low-light facial expression recognition |
| title_fullStr | A joint learning method for low-light facial expression recognition |
| title_full_unstemmed | A joint learning method for low-light facial expression recognition |
| title_short | A joint learning method for low-light facial expression recognition |
| title_sort | joint learning method for low light facial expression recognition |
| topic | Facial expression recognition Low-light image enhancement High-level vision task Cross high- and low-level learning Joint training |
| url | https://doi.org/10.1007/s40747-024-01762-z |
| work_keys_str_mv | AT yuanlunxie ajointlearningmethodforlowlightfacialexpressionrecognition AT jieou ajointlearningmethodforlowlightfacialexpressionrecognition AT bihanwen ajointlearningmethodforlowlightfacialexpressionrecognition AT zitongyu ajointlearningmethodforlowlightfacialexpressionrecognition AT wenhongtian ajointlearningmethodforlowlightfacialexpressionrecognition AT yuanlunxie jointlearningmethodforlowlightfacialexpressionrecognition AT jieou jointlearningmethodforlowlightfacialexpressionrecognition AT bihanwen jointlearningmethodforlowlightfacialexpressionrecognition AT zitongyu jointlearningmethodforlowlightfacialexpressionrecognition AT wenhongtian jointlearningmethodforlowlightfacialexpressionrecognition |