method: TencentAILab2017-10-28

Authors: Jingchao Zhou, Xing Ji, Zheng Zhou, Zhifeng Li

Description: Our method consists of two networks: text detection network employed in task 2.1 and word recognition network employed in task 2.2. Two networks are integrated with cascade training to achieve superior performance.

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.

Ranking Table

Description Paper Source Code
DateMethodRecallPrecisionHmean
2017-10-28TencentAILab97.20%96.30%96.74%
2018-01-22FOTS93.81%98.17%95.94%
2017-06-09TextBoxes++92.87%98.27%95.50%
2020-10-30MaskTextSpotter93.22%97.32%95.23%
2018-03-05HoText_v192.64%97.54%95.03%
2021-02-05AIX Lab., LG Electronics 92.76%96.13%94.41%
2017-03-20AdelaideMachineLearning91.24%97.14%94.10%
2016-08-23HUST_MCLAB90.77%97.25%93.90%
2017-07-11SRC-B-MachineLearningLab91.59%95.73%93.61%
2017-10-28MPTSys88.20%97.42%92.58%
2016-03-17Adelaide_ConvLSTMs84.93%98.91%91.39%
2015-04-03VGGMaxBBNet_09586.68%94.64%90.49%
2015-04-03VGGMaxBBNet (055)87.62%93.05%90.25%
2016-03-10SRC-B-TextProcessingLab84.46%95.13%89.48%
2017-01-19Yunos_Robot1.078.97%96.30%86.78%
2015-04-02StradVision-179.91%92.68%85.82%
2015-10-20Deep2Text II+75.82%96.29%84.84%
2015-04-01NJU Text (Version3)73.25%83.16%77.89%
2015-04-02Deep2Text II-272.08%83.49%77.37%
2015-04-01Deep2Text I69.74%85.78%76.93%
2015-04-02MSER MRF66.94%87.21%75.74%
2015-04-02Beam search CUNI (decoding for TextSpotter - no postprocessing)62.62%72.63%67.25%
2015-05-04Baseline OpenCV 3.0 + Tesseract52.92%77.97%63.05%
2015-04-02Beam search CUNI +S (decoding for TextSpotter - spell checked)16.47%93.38%28.00%
2015-04-01Baseline-TextSpotter0.00%0.00%0.00%

Ranking Graphic