MLe2e stands for Multi-Language end-to-end, which is used to evaluate the end-to-end reading system of scene text, as well as all intermediate stages such as text detection, script identification, and text recognition. The dataset contains a total of 711 scene images covering four different languages (Latin, Chinese, Kannada, and Korean).

Citation

@inproceedings{GomezDAS2016, author = {Lluis Gomez and Dimosthenis Karatzas}, title = {A fine-grained approach to scene text script identification}, booktitle = {DAS}, year = {2016}, }