Efficient and Accurate Arbitrary-Shaped Text Detection with Pixel Aggregation Network

TL;DR

This paper introduces the Pixel Aggregation Network (PAN), an efficient and accurate method for detecting arbitrary-shaped text in scenes, balancing speed and precision with a novel segmentation and post-processing approach.

Contribution

The paper proposes a new text detection framework combining a low-cost segmentation head and learnable post-processing for improved speed and accuracy.

Findings

Achieves 79.9% F-measure on CTW1500

Runs at 84.2 FPS, demonstrating real-time capability

Outperforms existing methods on standard benchmarks

Abstract

Scene text detection, an important step of scene text reading systems, has witnessed rapid development with convolutional neural networks. Nonetheless, two main challenges still exist and hamper its deployment to real-world applications. The first problem is the trade-off between speed and accuracy. The second one is to model the arbitrary-shaped text instance. Recently, some methods have been proposed to tackle arbitrary-shaped text detection, but they rarely take the speed of the entire pipeline into consideration, which may fall short in practical applications.In this paper, we propose an efficient and accurate arbitrary-shaped text detector, termed Pixel Aggregation Network (PAN), which is equipped with a low computational-cost segmentation head and a learnable post-processing. More specifically, the segmentation head is made up of Feature Pyramid Enhancement Module (FPEM) and Feature Fusion Module (FFM). FPEM is a cascadable U-shaped module, which can introduce multi-level information to guide the better segmentation. FFM can gather the features given by the FPEMs of different depths into a final feature for segmentation. The learnable post-processing is implemented by Pixel Aggregation (PA), which can precisely aggregate text pixels by predicted similarity vectors. Experiments on several standard benchmarks validate the superiority of the proposed PAN. It is worth noting that our method can achieve a competitive F-measure of 79.9% at 84.2 FPS on CTW1500.