Handwriting Recognition Using Deep Learning


In this project, we will be creating a model to recognize handwritten text in physical documents. Such a model would have applications in all industries where written text is still prevalant such as banking, insurance, and healthcare, as well as for other machine learning practitioners who wish to use handwritten text as data. This project will be a great learning opportunity in deep learning for any experience level as we will be exploring many deep learning algorithms such as fully connected, convolutional, and recurrent networks, as well as experimenting with state of the art computer vision algorithms (ResNet, UNet, etc). Additionally, the project has great entrepreneurial potential, being easily adapted into a webservice or a mobile app.

I am looking for developers who are eager to learn more about machine learning and who excel when working in a team. It is also necessary to have some programming background, preferably with Python. If you have any questions, feel free to send me an email at justinb.tran@mail.utoronto.ca.