Overview

Advanced topics in computer vision with a focus on the use of machine learning techniques and applications in graphics and human-computer interface. Topics include image representations, texture models, structure-from-motion algorithms, Bayesian techniques, object and scene recognition, tracking, shape modeling, and image databases. Applications may include face recognition, multimodal interaction, interactive systems, cinematic special effects, and photorealistic rendering. Covers topics complementary to 6.801/6.866; these subjects may be taken in sequence.

General information

Lecture: Mondays/Wednesdays 1:00-2:30pm
Room: 2-139 (where is this?)

Instructor: Antonio Torralba
E -mail: trrlb@mt.d (fill the missing vowels)
Office: D432

T.A.: Joseph Lim (office hours: Wednesday 3:30 - 4:30, office D428)

Material:

• Textbook: new book by Rick Szeliski (not published yet, but a draft is available online)
• Textbook: Computer vision: a modern approach, by Forsyth and Ponce. Prentice Hall, 2002.
• The class will make use of MATLAB.

Grading:

• problem sets: 1/3
• 2 take-home exams: 1/3
• final project: 1/3

Announcements

- Visit the course evaluation website (available until May 15).

- Gallery of pinhole and anaglyph images here.

Schedule

Resources

Related courses:

• Introduction to Computer Vision, by Michael Black
• Learning-Based Methods in Vision, by Alyosha Efros
• Computer Vision, by Kristen Grauman
• Computer Vision, by Rob Fergus
• Introduction to Computer Vision, by Fei-Fei Li

Other resources:

• The Computer Vision Industry

Code

Here there are links to useful code:

• The Steerable Pyramid
• Representation and Synthesis of Visual Texture, Portilla & Simoncelli
• Berkeley Segmentation
• Pb
• Superpixels
• Structure from Motion for Unordered Image Collections
• Peter Kovesi's Functions for Computer Vision
• SIFT implementation by Andrea Vedaldi
• Affine Covariant Features
• A simple object detector with boosting

Other useful code:

• Code for downloading Flickr images