EAST: An Efficient and Accurate Scene Text Detector

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.

PDF Abstract CVPR 2017 PDF CVPR 2017 Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Scene Text Detection COCO-Text EAST + VGG16 F-Measure 39.45 # 3
Precision 50.39 # 3
Recall 32.4 # 3
Scene Text Detection ICDAR 2015 EAST + PVANET2x RBOX (single-scale) F-Measure 78.2 # 35
Precision 83.6 # 36
Recall 73.5 # 39
Scene Text Detection ICDAR 2015 PAN F-Measure 82.9 # 29
Precision 84 # 35
Recall 81.9 # 24
Scene Text Detection MSRA-TD500 EAST + PVANET2x Recall 67.43 # 17
Precision 87.28 # 11
F-Measure 76.08 # 14
Scene Text Detection Total-Text EAST F-Measure 42.0% # 22
Precision 50.0 # 19
Recall 36.2 # 19


No methods listed for this paper. Add relevant methods here