Affiliation:
1. Laboratory of Signal Detection and Processing School of Information Science and Engineering, Xinjiang University Urumqi 830046 China
Abstract
AbstractScene text detection has been widely studied on haze‐free images with reliable ground truth annotation. However, detecting scene text in inclement weather conditions remains a major challenge due to the severe domain distribution mismatch problem. This paper introduces a domain adaptation curriculum learning method to address this problem. The scene text detector is self‐trained in an easy‐to‐hard manner using the pseudo‐labels predicted from foggy images. Thus, our method reduces the pseudo‐labeling noise level. Then, a feature alignment module is introduced to help the network learn domain‐invariant features by training a domain classifier. Experimental results show that our method improved significantly on both synthetic foggy data sets and natural foggy data sets, outperforming many state‐of‐the‐art scene text detectors. © 2024 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.
Reference18 articles.
1. WuW LuN XieE WangY YuW YangC Zhou H.Synthetic‐to‐real unsupervised domain adaptation for scene text detection in the wild.Proceedings of the Asian Conference on Computer Vision.2020.
2. SovianyP IonescuRT RotaP SebeN.Curriculum self‐paced learning for cross‐domain object detection.Computer Vision and Image Understanding.2021.
3. Unsupervised domain adaptation by backpropagation;Yaroslav G;International Conference on Machine Learning,2015
4. DFR‐TSD: a deep learning based framework for robust traffic sign detection under challenging weather conditions;Sabbir A;IEEE Transactions on Intelligent Transportation Systems,2021