Attention-based Feature Decomposition-Reconstruction Network for Scene Text Detection

TL;DR

This paper introduces an attention-based feature decomposition-reconstruction network that enhances scene text detection by leveraging contextual information and addressing over-segmentation, achieving state-of-the-art results.

Contribution

The novel network combines cross-level attention and feature decomposition-reconstruction modules to improve segmentation accuracy for arbitrary-shaped and large aspect ratio texts.

Findings

Achieves state-of-the-art performance on benchmark datasets.

Effectively reduces over-segmentation of large aspect ratio texts.

Enhances detection of curved and irregular text shapes.

Abstract

Recently, scene text detection has been a challenging task. Texts with arbitrary shape or large aspect ratio are usually hard to detect. Previous segmentation-based methods can describe curve text more accurately but suffer from over segmentation and text adhesion. In this paper, we propose attention-based feature decomposition-reconstruction network for scene text detection, which utilizes contextual information and low-level feature to enhance the performance of segmentation-based text detector. In the phase of feature fusion, we introduce cross level attention module to enrich contextual information of text by adding attention mechanism on fused multi-scaled feature. In the phase of probability map generation, a feature decomposition-reconstruction module is proposed to alleviate the over segmentation problem of large aspect ratio text, which decomposes text feature according to their frequency characteristic and then reconstructs it by adding low-level feature. Experiments have been conducted on two public benchmark datasets and results show that our proposed method achieves state-of-the-art performance.