Robust ReadingCompetition

method: Megvii-Image++2015-11-11

Authors: Cong Yao, Jianan Wu, Xinyu Zhou, Chi Zhang, Shuchang Zhou, Zhimin Cao, Qi Yin

Description: We train a Fully Convolutional Networks (FCN) to perform per-pixel prediction on the probability of text regions. Detections are formed by subsequent thresholding and partition operations in the prediction map.