PGNet: Real-time Arbitrarily-Shaped Text Spotting with Point Gathering Network

TL;DR

This paper introduces PGNet, a real-time, fully convolutional network for reading arbitrarily-shaped text that avoids complex post-processing and character annotations, achieving high speed and competitive accuracy.

Contribution

The paper proposes PGNet, a novel single-shot text spotting network that uses PG-CTC loss and a graph refinement module to improve efficiency and accuracy in reading irregular text.

Findings

Achieves 46.7 FPS on Total-Text, surpassing previous methods in speed.

Avoids NMS and RoI operations, simplifying the pipeline.

Maintains competitive accuracy in arbitrarily-shaped text reading.

Abstract

The reading of arbitrarily-shaped text has received increasing research attention. However, existing text spotters are mostly built on two-stage frameworks or character-based methods, which suffer from either Non-Maximum Suppression (NMS), Region-of-Interest (RoI) operations, or character-level annotations. In this paper, to address the above problems, we propose a novel fully convolutional Point Gathering Network (PGNet) for reading arbitrarily-shaped text in real-time. The PGNet is a single-shot text spotter, where the pixel-level character classification map is learned with proposed PG-CTC loss avoiding the usage of character-level annotations. With a PG-CTC decoder, we gather high-level character classification vectors from two-dimensional space and decode them into text symbols without NMS and RoI operations involved, which guarantees high efficiency. Additionally, reasoning the relations between each character and its neighbors, a graph refinement module (GRM) is proposed to optimize the coarse recognition and improve the end-to-end performance. Experiments prove that the proposed method achieves competitive accuracy, meanwhile significantly improving the running speed. In particular, in Total-Text, it runs at 46.7 FPS, surpassing the previous spotters with a large margin.