ECSE 626 — Statistical Computer Vision

An overview of statistical and machine learning techniques as applied to computer vision problems, including: stereo vision, motion estimation, object and face recognition, image registration and segmentation. Topics include regularization, probabilistic inference, information theory, Gaussian Mixture Models, Markov-Chain Monte Carlo methods, importance sampling, Markov random fields, principal and independent components analysis, probabilistic deep learning methods including variational models, Bayesian deep learning.

• Rating: 3.00 out of 5 from 38 student reviews
• Difficulty: 4.00 out of 5
• Credits: 4
• Faculty: Graduate Studies
• Department: Electrical & Computer Engr
• Taught by: Tal Arbel
• Prerequisites: (ECSE 205 or equivalent) and (ECSE 415 or COMP 558 or equivalent).

Sections offered

• Section 001 (Lec), Mon Wed 4:05-5:25 pm — 10 seats open