ViSA: Visual and Semantic Alignment for Robust Scene Text Recognition.

Zhenru Pan, Zhilong Ji, Xiao Liu, Jinfeng Bai, Cheng-Lin Liu

Browse the full ICDAR paper archive.