ECE 59500 - Deep Learning for Computer Vision

Course Details

Lecture Hours: 3 Credits: 3

Areas of Specialization:

  • Computer Engineering

Counts as:

  • EE Elective
  • CMPE Selective - Special Content

Normally Offered:

Each Fall


On-campus only


ECE 26400 Minimum grade of B, ECE 36900 Minimum grade of B, ECE30200 Minimum grade of B, MA 26200 or MA 26500 Minimum grade of B, ECE 60000 for graduate students.

Requisites by Topic:

Advanced C Programming, Discrete Math, Probability and Linear Algebra

Catalog Description:

An introduction to modern computer vision using methods from machine learning and deep learning. Covers segmentation, object classification and localization, activity classification and localization, semantic segmentation, depth reconstruction, 3D reconstruction, generative adversarial networks, image and video captioning, and image and video retrieval. The course will cover fundamental topics as well as recent advances from the literature.

Required Text(s):

  1. Computer Vision: A Modern Approach , 2nd Edition , David A. Forsyth and Jean Ponce Pearson , ISBN No. 13: 978-0136085928

Recommended Text(s):

  1. Deep Learning , Ian Goodfellow, Yoshua Bengio, and Aaron Courville , MIT Press , ISBN No. 13: 978- 0262035613

Learning Outcomes

A student who successfully fulfills the course requirements will have demonstrated:

  • Ability to design and implement an object classifier and localizer
  • ability to design and implement an activity classifier and localizer
  • ability to design and implement an image or object captioning system
  • ability to present research results in CV to peers

Lecture Outline:

Lectures Lecture Topics
Topic Fundamentals of neural networks and deep learning
Topic Graph-based image segmentation
Topic Object classification and localization
Topic LSTMs
Topic Activity classification and localization
Topic Semantic segmentation
Topic depth reconstruction
Topic 3D reconstruction
Topic generative adversarial networks
Topic Image and video captioning
Topic Image and video retrieval

Assessment Method:

Homework, reports, projects and presentations