EAST: An Efficient and Accurate Scene Text Detector

TL;DR

EAST introduces a streamlined scene text detection pipeline that directly predicts oriented text regions in images, achieving superior accuracy and speed compared to previous methods.

Contribution

The paper presents a simple, single neural network pipeline for scene text detection that eliminates intermediate steps, improving both efficiency and accuracy.

Findings

Achieves an F-score of 0.7820 on ICDAR 2015

Runs at 13.2 fps on 720p images

Outperforms state-of-the-art methods in accuracy and speed

Abstract

Previous approaches for scene text detection have already achieved promising performances across various benchmarks. However, they usually fall short when dealing with challenging scenarios, even when equipped with deep neural network models, because the overall performance is determined by the interplay of multiple stages and components in the pipelines. In this work, we propose a simple yet powerful pipeline that yields fast and accurate text detection in natural scenes. The pipeline directly predicts words or text lines of arbitrary orientations and quadrilateral shapes in full images, eliminating unnecessary intermediate steps (e.g., candidate aggregation and word partitioning), with a single neural network. The simplicity of our pipeline allows concentrating efforts on designing loss functions and neural network architecture. Experiments on standard datasets including ICDAR 2015, COCO-Text and MSRA-TD500 demonstrate that the proposed algorithm significantly outperforms state-of-the-art methods in terms of both accuracy and efficiency. On the ICDAR 2015 dataset, the proposed algorithm achieves an F-score of 0.7820 at 13.2fps at 720p resolution.