Fused Text Segmentation Networks for Multi-oriented Scene Text Detection

TL;DR

This paper introduces Fused Text Segmentation Networks, a novel end-to-end framework that combines semantic segmentation and object detection techniques for accurate multi-oriented scene text detection, outperforming existing methods on key benchmarks.

Contribution

The paper proposes a new unified network architecture that effectively fuses multi-level features for joint text detection and segmentation without extra pipelines.

Findings

Achieves state-of-the-art results on ICDAR2015 and MSRA-TD500 benchmarks.

Surpasses previous methods in multi-oriented scene text detection accuracy.

Demonstrates effectiveness on curved text in the total-text dataset.

Abstract

In this paper, we introduce a novel end-end framework for multi-oriented scene text detection from an instance-aware semantic segmentation perspective. We present Fused Text Segmentation Networks, which combine multi-level features during the feature extracting as text instance may rely on finer feature expression compared to general objects. It detects and segments the text instance jointly and simultaneously, leveraging merits from both semantic segmentation task and region proposal based object detection task. Not involving any extra pipelines, our approach surpasses the current state of the art on multi-oriented scene text detection benchmarks: ICDAR2015 Incidental Scene Text and MSRA-TD500 reaching Hmean 84.1% and 82.0% respectively. Morever, we report a baseline on total-text containing curved text which suggests effectiveness of the proposed approach.