TextSnake: A Flexible Representation for Detecting Text of Arbitrary Shapes

TL;DR

TextSnake introduces a flexible, geometry-based representation for scene text detection that effectively captures arbitrary-shaped text, including curved text, achieving state-of-the-art results on multiple benchmarks.

Contribution

The paper proposes TextSnake, a novel representation for scene text detection that models text as a sequence of overlapping disks, enabling detection of arbitrary-shaped text.

Findings

Achieves state-of-the-art performance on Total-Text and SCUT-CTW1500 datasets.

Outperforms baseline by over 40% in F-measure on Total-Text.

Effectively detects curved and arbitrarily shaped text in natural images.

Abstract

Driven by deep neural networks and large scale datasets, scene text detection methods have progressed substantially over the past years, continuously refreshing the performance records on various standard benchmarks. However, limited by the representations (axis-aligned rectangles, rotated rectangles or quadrangles) adopted to describe text, existing methods may fall short when dealing with much more free-form text instances, such as curved text, which are actually very common in real-world scenarios. To tackle this problem, we propose a more flexible representation for scene text, termed as TextSnake, which is able to effectively represent text instances in horizontal, oriented and curved forms. In TextSnake, a text instance is described as a sequence of ordered, overlapping disks centered at symmetric axes, each of which is associated with potentially variable radius and orientation. Such geometry attributes are estimated via a Fully Convolutional Network (FCN) model. In experiments, the text detector based on TextSnake achieves state-of-the-art or comparable performance on Total-Text and SCUT-CTW1500, the two newly published benchmarks with special emphasis on curved text in natural images, as well as the widely-used datasets ICDAR 2015 and MSRA-TD500. Specifically, TextSnake outperforms the baseline on Total-Text by more than 40% in F-measure.