Bidirectional Regression for Arbitrary-Shaped Text Detection

Abstract

Arbitrary-shaped text detection has recently attracted increasing interests and witnessed rapid development with the popularity of deep learning algorithms. Nevertheless, existing approaches often obtain inaccurate detection results, mainly due to the relatively weak ability to utilize context information and the inappropriate choice of offset references. This paper presents a novel text instance expression which integrates both foreground and background information into the pipeline, and naturally uses the pixels near text boundaries as the offset starts. Besides, a corresponding post-processing algorithm is also designed to sequentially combine the four prediction results and reconstruct the text instance accurately. We evaluate our method on several challenging scene text benchmarks, including both curved and multi-oriented text datasets. Experimental results demonstrate that the proposed approach obtains superior or competitive performance compared to other state-of-the-art methods, e.g., 83.4% F-score for Total-Text, 82.4% F-score for MSRA-TD500, etc.

arXiv commentAccepted at ICDAR 2021, 15 pages

References in corpus (5)

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• TextBoxes++: A Single-Shot Oriented Scene Text Detector
• TextBoxes: A Fast Text Detector with a Single Deep Neural Network
• TextField: Learning A Deep Direction Field for Irregular Scene Text Detection
• Detecting Curve Text in the Wild: New Dataset and New Solution