Deep structured output learning for unconstrained text recognition

We develop a representation suitable for the unconstrained recognition of words in natural images: the general case of no fixed lexicon and unknown length. To this end we propose a convolutional neural network (CNN) based architecture which incorporates a Conditional Random Field (CRF) graphical mod...

Mô tả đầy đủ

Chi tiết về thư mục
Những tác giả chính: Jaderberg, M, Simonyan, K, Vedaldi, A, Zisserman, A
Định dạng: Conference item
Ngôn ngữ:English
Được phát hành: International Conference on Learning Representations 2015