Course Information

We live in a world that spans 3 dimensions. Cameras and sensors image the 3D world by projecting to a 2D plane. How can we recover the 3D world back from these images? What techniques can we use to process 3D data? In this course we will study computer vision and machine learning techniques to recover 3D information of the world from images, and process and understand 3D data. We will learn about classical computer vision techniques but focus on cutting-edge deep learning methods. The techniques we will study are widely used, for instance, in self-driving cars and smartphone AR face filter apps.

Class Time: Tu Th, 9:00-10:20 AM, CIT 241
Email: srinath@brown.edu
Ed Stem: https://edstem.org/us/courses/74668
Capstone Request: Google Form

News

• Class starts on Thursday, Jan 31 in CIT 241.
• Please register/request override directly on Courses @ Brown with a short note on how you meet the recommended background preparation.

Learning Goals

This course has two main learning goals. Students are expected to actively participate in class including discussions and group activities.

1. Learn about the state of the art in 3D computer vision and machine learning. We will do this by reading a curated list of research papers on relevant topics.
2. Understand research practice in computer science, with specific focus on the computer vision and ML communities. We will learn how to effectively read papers, write reviews, present papers, critique and discuss research, and do a group research project.

Recommended Background

This is an advanced course but students at all levels are welcome to participate if they have the necessary background. We recommend that you take one or more of the following courses or their equivalents before enrolling.

• CSCI 1230 Introduction to Computer Graphics
• CSCI 1420 Machine Learning
• CSCI 1430 Computer Vision
• ENGN 1610 Image Understanding
• CSCI 1470 Deep Learning