MEAN: Multi - Element Attention Network for Scene Text Recognition.
2020 25th International Conference on Pattern Recognition (ICPR)（2021）
关键词
attention weights,multielement attention network,feature extraction,MEA mechanism,text codes,orientational positional encoding,feature maps output,feature vector sequence,seven public English scene text datasets,RCTW Chinese scene text dataset,irregular scene text samples,multi- element attention network,scene text recognition,image quality,text instances,natural scene images,scene texts,novel multielement attention mechanism,global levels,scene text image,self-attention technique
AI 理解论文
溯源树
样例
生成溯源树，研究论文发展脉络
Chat Paper
正在生成论文摘要