Basic Information
The Goal of This Class Is to Expose The Students to a Variety Of Modern Statistical Methods For Solving Image Processing Problems, With Emphasis On Generative Models. The Course Will Cover Methods For Learning Distributions and Their Applications in Image Processing. Those Include Nonparametric Methods, Invertible Netwroks, Generative Adversarial Networks (gans), Markov Random Fields, Mcmc Based Sampling Methods, Methods For Training Energy Based Models (ebms), Introduction to Diffusion Based Generative Models, Applications To Image Restoration, Editing and Generation. Learning Outcomes# at The End of The Course The Students Will# 1. Gaining Familiarity With Basic Statistical Methods. 2. Use In Image Processing. The Course Successfully Will Be Able to Fit The Statistical Models 3. Use Them For Solving a Variety of Problems.
Faculty: Electrical and Computer Engineering
|Graduate Studies
Pre-required courses
(44202 - Random Signals and 46195 - Machine Learning and 46200 - Image Processing and Analysis)