Adaptive Segmentation Network for Scene Text Detection
This work addresses challenges in scene text detection for computer vision applications, representing an incremental improvement by refining existing segmentation-based methods.
The paper tackles the problem of threshold selection bottlenecks and poor performance on text instances with extreme aspect ratios in scene text detection by proposing an Adaptive Segmentation Network (ASNet) that automatically learns segmentation thresholds and uses a Global-information Enhanced Feature Pyramid Network, achieving state-of-the-art performance on four benchmarks including ICDAR 2015 and MSRA-TD500.
Inspired by deep convolution segmentation algorithms, scene text detectors break the performance ceiling of datasets steadily. However, these methods often encounter threshold selection bottlenecks and have poor performance on text instances with extreme aspect ratios. In this paper, we propose to automatically learn the discriminate segmentation threshold, which distinguishes text pixels from background pixels for segmentation-based scene text detectors and then further reduces the time-consuming manual parameter adjustment. Besides, we design a Global-information Enhanced Feature Pyramid Network (GE-FPN) for capturing text instances with macro size and extreme aspect ratios. Following the GE-FPN, we introduce a cascade optimization structure to further refine the text instances. Finally, together with the proposed threshold learning strategy and text detection structure, we design an Adaptive Segmentation Network (ASNet) for scene text detection. Extensive experiments are carried out to demonstrate that the proposed ASNet can achieve the state-of-the-art performance on four text detection benchmarks, i.e., ICDAR 2015, MSRA-TD500, ICDAR 2017 MLT and CTW1500. The ablation experiments also verify the effectiveness of our contributions.