Description: Recognition model: Based on Transformer with backbone ResNet50. A voting process is done to identify the language of recognized transcript. Train-set: 2017 MLT task2 train-set & 2019 MLT task2 train-set & 2019 MLT Synthetic dataset.
Authors: Changxu Cheng
Description: Convolution layers followed by 2 branches: Global Squeezer (GS) and Patch Aggregator (PA). Other process is the same as in HUST (GS). The paper will be available soon.
Authors: Sunghyo Chung, Youngmin Baek, Hwalsuk Lee, Jaegul Choo
Description: We formulate script identification as a semantic segmentation task. We use both MLT task1 and task2 datasets for training. CLOVA-AI team, Naver Corp.