SegLink++: Detecting Dense and Arbitrary-shaped Scene Text by Instance-aware Component Grouping. (December 2019)

• Record Type: Journal Article
• Authors: Tang, Jun; Yang, Zhibo; Wang, Yongpan; Zheng, Qi; Xu, Yongchao; Bai, Xiang
• Is Part Of: Pattern recognition. Volume 96(2019:Dec.)
• Publication Date: 2019-12
• Subjects: Scene text detection -- Multi-oriented text -- Curve text -- Dense text; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2019.06.020
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: A novel bottom-up method for detecting dense and arbitrary-shaped scene texts. Explicitly learning repulsive links between close texts helps to separate dense text instances. Instance-aware loss for bottom-up deep learning-based methods, further boosting the performance. A dataset consisting of dense and arbitrary-shaped scene text of commodity images is introduced. Significantly improved performance on dense and curved text detection. Abstract: State-of-the-art methods have achieved impressive performances on multi-oriented text detection. Yet, they usually have difficulty in handling curved and dense texts, which are common in commodity images. In this paper, we propose a network for detecting dense and arbitrary-shaped scene text by instance-aware component grouping (ICG), which is a flexible bottom-up method. To address the difficulty in separating dense text instances faced by most bottom-up methods, we propose attractive and repulsive link between text components which forces the network learning to focus more on close text instances, and instance-aware loss that fully exploits context to supervise the network. The final text detection is achieved by a modified minimum spanning tree (MST) algorithm based on the learned attractive and repulsive links. To demonstrate the effectiveness of the proposed method, we introduce a dense and arbitrary-shaped scene text dataset composed of commodity images (DAST1500). Experimental results show that the proposed ICG significantly outperforms state-of-the-art methods on DAST1500 and two curved text datasets: Total-Text and CTW1500, and also achieves very competitive performance on two multi-oriented datasets: ICDAR15 (at 7.1FPS for 1280 × 768 image) and MTWI.