Generative Image Synthesis with Deep Learning

Generative Image Synthesis with Deep Learning

Recent studies in Deep Learning have led to many advancements in problems regarding images. One of the obstacles that researchers have faced is training with Deep Learning methods requires a vast amount of data and computation power. Among the methods that try to overcome with mentioned problem, we are especially interested in transfer learning and using synthetic data.
Our goal is to find a mapping between images of characters of digital fonts and handwritten images of same characters. By making use of this method, our aim is to be able to generate handwritten sentences. We also would like to try style transfer later to capture the handwriting of user with interactive tools and generate indistinguishable handwritten texts.

Project Poster: 

Project Members: 

Enis Simsar
Ömer Kırbıyık

Project Advisor: 

Ali Taylan Cemgil

Project Status: 

Project Year: 

  • Fall

Contact us

Department of Computer Engineering, Boğaziçi University,
34342 Bebek, Istanbul, Turkey

  • Phone: +90 212 359 45 23/24
  • Fax: +90 212 2872461

Connect with us

We're on Social Networks. Follow us & get in touch.