PhDinfo seminars: Zarah Ziran on "Document Analysis by Deep Learning"

PhDinfo seminars: Zarah Ziran on "Document Analysis by Deep Learning"

by SCHOEN FABIO -
Number of replies: 0

A reminder: tomorrow at 14 in room 107 (via di S Marta 3 Firenze) Zarah will give a talk on "Document Analysis by Deep Learning"


Abstract



The technology of document analysis and recognition (DAR) is to analyze 
the structure and textual contents of document images and handwriting. 
It faces numerous application needs such as digitization of books and 
forms, pen-based text input, information extraction from Web document 
images. It has been under study as a field of pattern recognition since 
1960s. In recent years, the introduction of deep learning to DAR has led 
to significant improvement of performance in many branches, particularly 
in the cases when large sets of labeled data are available for 
supervised learning, such as handwritten character and text recognition. 
Among the most successful deep learning models are the convolutional 
neural network (CNN) and the recurrent neural network with long 
short-term memory (LSTM). The application of deep learning is now 
extended to scene text detection and recognition, document image 
segmentation and layout analysis, writer identification, document 
retrieval, and so on.