For the course project you will explore a topic in-depth of your own choosing. This can be an implementation (implement an existing algorithm); an application (apply a computer vision algorithm to a new problem); or research (trying to invent something new).

To get you started, we have prepared a list of suggested projects. We believe that any of these would be feasible to complete as a project.

We expect you to work in groups of 3-5 students for the course project. The course project should amount to roughly one homework worth of work per person. In previous years we have typically expected about two homework assignments worth of work per person in the group project; however we are explicitly lowering our expectations to account for the extra overhead of collaborating remotely.

We are not expecting state-of-the-art, publication ready results from your course project! The point of the project is to get practice applying concepts of the class to a problem of your choosing, without the “scaffolding” code provided in the homeworks.

What to submit

The project proposal is due on Monday, April 5, 2021 11:59:59pm to Gradescope.
You only need to submit one project propsal per group and add all the other members on Gradescope.

After submitting your project proposal, please fill out this Google Form once as a group to help us keep track of who is working together.

Your project proposal should be a 1-page PDF that answers the following questions:

Project Title: What is the name of your project?

Group Members: What are the names and uniqnames of the students involved?

Problem Statement: What is the problem you are trying to solve?

Approach: How do you plan to go about solving this problem? You don’t need to have everything figured out exactly, but you should have a vague sense of how you will proceed.

Data: What dataset do you plan to use? A common failure mode for projects is to have a cool idea, but no idea where to get the necessary data. We recommend against collecting your own dataset for the project, as this will significantly increase the complexity and workload; instead you should try to get away with existing datasets.

Computational Resources: What computational resources will you use for this project? For some projects a laptop may be completely fine. But if you are planning to train any kind of neural network, you should have an estimate of how much time a model should take to train, and where you will get access to the computational resources you need. Google Colab is a great free resources for small amounts of GPU resources; but be aware that this is not sufficient for training large-scale models.

Evaluation: How do you plan to evaluate whether your project is successful? What metric will you use? Is there some simple baseline that you plan to compare your model against?

If you are following one of our suggested projects then your “Problem Statement” can be very brief – one or two sentences is fine. For the suggested projects, we don’t expect you to work on all of the datasets we link; but please do tell us which you are planning to use.