Focus-Enhanced Scene Text Recognition with Deformable Convolutions

TL;DR

This paper introduces a focus-enhanced scene text recognition method using deformable convolutions, effectively handling irregular and distorted text without explicit rectification, and demonstrates strong performance on benchmark datasets.

Contribution

The work proposes a novel recognition network that leverages deformable convolutions to adapt to irregular text shapes without rectification steps.

Findings

Achieves state-of-the-art results on public benchmarks.

Effectively recognizes irregular and distorted text.

Demonstrates the effectiveness of deformable convolutions in scene text recognition.

Abstract

Recently, scene text recognition methods based on deep learning have sprung up in computer vision area. The existing methods achieved great performances, but the recognition of irregular text is still challenging due to the various shapes and distorted patterns. Consider that at the time of reading words in the real world, normally we will not rectify it in our mind but adjust our focus and visual fields. Similarly, through utilizing deformable convolutional layers whose geometric structures are adjustable, we present an enhanced recognition network without the steps of rectification to deal with irregular text in this work. A number of experiments have been applied, where the results on public benchmarks demonstrate the effectiveness of our proposed components and shows that our method has reached satisfactory performances. The code will be publicly available at https://github.com/Alpaca07/dtr soon.