SFBigAnalytics_20210601: Full Page Handwriting Recognition via Image to Sequence Extraction

preview_player
Показать описание
In this talk, Sumeet Singh i will present a Neural Network based Handwritten Text Recognition (HTR) model architecture that can be trained to recognize full pages of handwritten or printed text without segmentation of images into lines or paragraphs (as prior art requires). I will talk about the challenges in addressing this problem, review the field, and provide details about our new state of art full page text recognition model. The model is also very easy to replicate using off the shelf modules, and therefore should be of particular interest to ML engineers wanting to develop a state-of-art handwriting recognition system.

Speaker: Sumeet Singh (Turnitin)

Sumeet Singh is a Distinguished ML Scientist at Turnitin focussed on applied research in Computer Vision, NLU and AI in general.
Комментарии
Автор

can u tell me, where can I get the dataset?

nurayuff
Автор

Thanks for the video. Where can I get the Code for the works done? Is it applicable for urdu database

azeemece