Skip to Main Content

COMPSCI280A

Download as PDF

Intro to Computer Vision and Computational Photography

Electrical Engineering and Computer SciencesGraduateCOE - College of Engineering
Overview

Subject

COMPSCI

Course Number

280A

Department

Electrical Engineering and Computer Sciences

Course Level

Graduate

Course Title

Intro to Computer Vision and Computational Photography

Course Description

This course introduces students to computing with visual data (images and video). We will cover acquisition, representation, and manipulation of visual information from digital photographs (image processing), image analysis and visual understanding (computer vision), and image synthesis (computational photography). Key algorithms will be presented, ranging from classical to contemporary, with an emphasis on using these techniques to build practical systems. The hands-on emphasis will be reflected in the programming assignments, where students will acquire their own images and develop, largely from scratch, image analysis and synthesis tools for real-world applications.

Minimum

4

Maximum

4

Grading Basis

Default Letter Grade; S/U Option

Instructors

Efros, Kanazawa

American Cultures Requirement

No

Reading and Composition Requirement

None of the Reading and Composition Requirement

Repeat Rules

Course is not repeatable for credit.

Credit Replacement Courses. Upon passing, students can use the following course(s) to replace a deficient grade for this course.

-

Course Objectives

Students will learn classic algorithms in image manipulation with Gaussian and Laplacian Pyramids, understand the hierarchy of image transformations including homographies, and how to warp an image with these transformations., Students will learn how to apply Convolutional Neural Networks for computer vision problems and how they can be used for image manipulation. mechanics of a pin-hole camera, representation of images as pixels, physics of light and the process of image formation, to manipulating the visual information using signal processing techniques in the spatial and frequency domains. Students will learn the fundamentals of 3D vision: stereo, multi-view geometry, camera calibration, structure-frommotion, multi-view stereo, and the plenoptic function

Student Learning Outcomes

After this class, students will be comfortable implementing, from scratch, these algorithms in modern programming languages and deep learning libraries.

Formats

Discussion, Lecture

Term

Fall and Spring

Duration (in weeks)

15

Minimum Hours

3

Maximum Hours

3

Lecture Mode of Instruction

In Person

Minimum Hours

1

Maximum Hours

1

Discussion Mode of Instruction

In Person

Outside Work Hours Min

10

Maximum Hours

10
Powered by Coursedog