method: FOTS2018-01-22

Authors: Xuebo Liu, Ding Liang, Shi Yan, Dagui Chen, Yu Qiao, Junjie Yan

Description: A unified end-to-end trainable Fast Oriented Text Spotting (FOTS) network for simultaneous detection and recognition, sharing computation and visual information among the two complementary tasks.

method: CRAFT2019-04-08

Authors: Youngmin Baek, Bado Lee, Dongyoon Han, Sangdoo Yun, and Hwalsuk Lee

Description: We propose a novel text detector called CRAFT. The proposed method effectively detects text area by exploring each character and affinity between characters. To overcome the lack of individual character level annotations, our framework exploits the pseudo character-level bounding boxes acquired by the learned interim model in a weakly-supervised manner.

Clova AI OCR Team, NAVER/LINE Corp.

Ranking Table

Description Paper Source Code
DateMethodRecallPrecisionHmean
2018-01-22FOTS87.92%91.85%89.84%
2019-04-08CRAFT84.26%89.79%86.93%
2019-05-03Mask TextSpotter81.70%90.80%86.01%
2017-09-13PixelLink83.77%86.65%85.19%
2017-07-26TextBoxes++80.79%89.11%84.75%
2018-01-04crpn80.69%88.77%84.54%
2019-01-20east_credit78.14%88.74%83.10%
2019-04-10EAST-VGG1681.27%84.36%82.79%
2017-01-23RRPN-477.13%83.52%80.20%
2019-03-08R2CNN++ (single scale)78.86%81.33%80.08%
2016-10-28RRPN-373.23%82.17%77.44%

Ranking Graphic