TY - GEN
T1 - FPN-Retinex Enables Semi-Supervised Low-Light Image Enhancement Via Feature Pyramid Network with Flexible Backbones
AU - Mao, Ruiqi
AU - Xu, Hui
AU - Cui, Rongxin
N1 - Publisher Copyright:
© 2024 Technical Committee on Control Theory, Chinese Association of Automation.
PY - 2024
Y1 - 2024
N2 - Nowadays unsupervised low-light image enhancement methods reduce the demand for training data and the risk of overfitting. Yet they face challenges in achieving a balance between model performance and efficiency when handling real-world low-light images in unknown complex scenarios. Herein, we present a semi-supervised low-light image enhancement scheme termed FPN-Retinex that can leverage the supervision of a limited number of low-light/normal image pairs to realize an accurate Retinex decomposition, and based on this, achieve brightening the illumination of unpaired images to reduce dependence on paired datasets and improve generalization ability. The decomposition network is learned with some newly established constraints for complete decoupling between reflectance and illumination. For the first time, we introduce the feature pyramid network (FPN) to adjust the illumination maps of other low-light images without any supervision. Under this flexible framework, a wide range of backbones can be employed to work with illumination map generator, to navigate the balance between performance and efficiency. In addition, a novel attention mechanism is integrated into the FPN for giving the adaptability towards application scenes with different environment like underwater image enhancement (UIE) and dark face detection. Extensive experiments demonstrate that our proposed scheme has a more robust performance with high efficiency facing various images from different low-light environments over state-of-the-art methods.
AB - Nowadays unsupervised low-light image enhancement methods reduce the demand for training data and the risk of overfitting. Yet they face challenges in achieving a balance between model performance and efficiency when handling real-world low-light images in unknown complex scenarios. Herein, we present a semi-supervised low-light image enhancement scheme termed FPN-Retinex that can leverage the supervision of a limited number of low-light/normal image pairs to realize an accurate Retinex decomposition, and based on this, achieve brightening the illumination of unpaired images to reduce dependence on paired datasets and improve generalization ability. The decomposition network is learned with some newly established constraints for complete decoupling between reflectance and illumination. For the first time, we introduce the feature pyramid network (FPN) to adjust the illumination maps of other low-light images without any supervision. Under this flexible framework, a wide range of backbones can be employed to work with illumination map generator, to navigate the balance between performance and efficiency. In addition, a novel attention mechanism is integrated into the FPN for giving the adaptability towards application scenes with different environment like underwater image enhancement (UIE) and dark face detection. Extensive experiments demonstrate that our proposed scheme has a more robust performance with high efficiency facing various images from different low-light environments over state-of-the-art methods.
KW - adversarial learning
KW - feature pyramid network
KW - low-light image enhancement
KW - pretrained backbones
KW - Retinex decomposition
KW - semi-supervised learning
UR - http://www.scopus.com/inward/record.url?scp=85205456313&partnerID=8YFLogxK
U2 - 10.23919/CCC63176.2024.10662161
DO - 10.23919/CCC63176.2024.10662161
M3 - 会议稿件
AN - SCOPUS:85205456313
T3 - Chinese Control Conference, CCC
SP - 7325
EP - 7332
BT - Proceedings of the 43rd Chinese Control Conference, CCC 2024
A2 - Na, Jing
A2 - Sun, Jian
PB - IEEE Computer Society
T2 - 43rd Chinese Control Conference, CCC 2024
Y2 - 28 July 2024 through 31 July 2024
ER -