UROP Openings

Have a UROP opening you would like to submit?

Please fill out the form.

Submit your UROP opening

Applied Computer Vision in the Wild


Term:

Fall

Department:

CSAIL: Computer Science and Artificial Intelligence Lab

Faculty Supervisor:

Pulkit Agrawal

Faculty email:

pulkitag@mit.edu

Apply by:

Sep 15 2020

Contact:

To apply use this form: https://forms.gle/ywNrXD8he5vqZwPP9

Project Description

Despite the recent advances in using deep learning models for computer-vision tasks, there exist various factors that cause our trained deep visual systems to fail in the real-world. These alarming discrepancies between perfectly curated images used in training datasets and the imperfect images we encounter in the real world have not been addressed. To better understand the limitations of current deep computer vision models, we are interested in creating a standardized framework to test out state-of-the-art algorithms in real-world settings, understand the potential shortcomings of these models, and come up with solutions that could alleviate these issues. Find more about our lab here: https://people.csail.mit.edu/pulkitag/

Pre-requisites

We are looking for someone who has background knowledge in computer vision and machine learning, has mathematical and programming maturity and is comfortable using deep-learning frameworks such as PyTorch.