Robust ReadingCompetition
Challenges

method: SCUT-DLVClab22017-06-30

Authors: Yuliang Liu, Canjie Luo, Lianwen Jin, Sheng Zhang, Zhaohai Li, Lele Xie, Zenghui Sun

Description: Two models have been trained separately: one model for text detection and another for classifying scripts. The two models are jointed to output the final results. After generating the detection results, a classification model with 8 classes (including background) is used to discard the detected boxes classified as background with very high confidence. Then a 7-class model was utilized to yield the final results. Since Chinese and Japanese are found to be easily confused by the model, and since only few images simultaneously contain both Chinese and Japanese scripts, a statistical average method is used to modify the Chinese and Japanese classification results.

method: TH-DL2017-06-30

Authors: Donglai Xiang, Jiaming Guo, Guangxiang Bin, Liangrui Peng, Changsong Liu

Description: The method is an integration of the methods of Tasks 1 and 2 (see their descriptions above under the same method name). All the methods are implemented by PyTorch.

Ranking Table

Description Paper Source Code
DateMethodAverage PrecisionHmeanPrecisionRecall
2017-06-30SCUT-DLVClab241.42%58.08%71.78%48.77%
2017-06-30TH-DL24.54%39.37%58.58%29.65%

Ranking Graphic

Ranking Graphic