Laboratory Learning Program 2023 - Exploratory Computer Vision Project (COS-01)

Research Opportunity Number: C0S-01

Project Title:  Exploratory Computer Vision Project

Project Summary: Computer vision is developing algorithms and software to better understand the world primarily through images and videos. In the first half, students will explore existing computer vision works in a specific task of their choice (e.g. classification, segmentation, image captioning) and develop an analysis of strengths and limitations of the current methods. In the second half, they can select one of the existing works and experiment with it to achieve better performance. The students will learn skills including reading academic papers, training machine learning models using Pytorch, running experiments, and writing code to analyze results. They may also attend relevant faculty lectures at Princeton AI4ALL (http://ai4all.princeton.edu). The internship will be adapted to the students’ unique skill sets and interests. Students from groups historically underrepresented in computer science are especially encouraged to apply.

IMPORTANT: All applicants must also fill out this Google form to be considered  https://docs.google.com/forms/d/e/1FAIpQLSc1AaxNiVspY4xCgnwjfxxjzybZCz9iLhh0hcVAjqSDvW3ltg/viewform

Student Roles and Responsibilities: The students will work on an independent research project under supervision of graduate student(s) and faculty advisor; learn to use libraries like Pytorch in computer vision research; ask lots of questions, present findings of the project at the end of the internship to the lab group.

Additional Considerations:  Start and end dates are flexible, but the student should be available essentially full-time (at least 35 hours/week) for a 4-5 week commitment.

Department/Institute:  Computer Science

Faculty Sponsor:   Olga Russakovsky

Participation Dates: July 5 - August 18, 2023

Stipend Offered:  $300

Number of Internships Available: 0-2

Application Deadline: March 15, 2023, midnight eastern daylight time