CVPR2019論文サマリ

tag: video-question-answering

ESIR: End-To-End Scene Text Recognition via Iterative Image Rectification
by: kota yoshida
video question answering video