Deliverables and grading policy
You will have the following deliverables through the course:
Assignments 1 and 2 (20% each)
The assignments have to be done individually. However, you can discuss with each other.
Submit a text file along with the assignment naming those you discuss with.
Paper summaries and reviews (10%)
This fall, I am including a new set of deliverables: a set of paper reviews.
Being able to read a paper and grasp both what the paper is about, and how it falls short is a crucial research skill no matter what area you work in.
This has to be done individually.
The course schedule lists some readings for certain lectures.
Before the lecture, you are required to read the paper and submit a short review.
This review should consist of two paragraphs.
The first paragraph should be a summary of the paper.
The second paragraph should be a critique: it should talk about one way in which the paper could be improved or extended.
Again, the review must be submitted before the corresponding lecture
Project (50%)
The biggest requirement of the course is a final project.
This project should be a research project, in the sense that it should create new knowledge: it should tell the broader community something that they did not know before.
However, the scope of this question need not be at the level of a full-fledged paper.
Indeed, I would recommend that you identify a research project that you think is doable in the time provided.
Here are some kinds of projects that you might do:
- How do I design a new approach for problem X? (Think about what is missing in current approaches)
- How does method A compare to method B for problem C? (Assuming this comparison has not been done before)
- How does method A compare to method B in new evaluation metric E? (Think of why this evaluation metric makes sense, and what insight you will gain from this metric)
- Why does method A work / not work for problem B?
If your research area is not computer vision, you can do projects that either apply ideas from your research area to computer vision, or vice versa.
The core requirement of producing new knowledge for your field still stands.
PhD students are encouraged to do projects in their research area. You can use the same project for multiple classes provided it is of a correspondingly large scope.
The project itself will have three main deliverables: the project proposal, the final report and a project presentation.
The project should be done in groups of 2-4. If you are doing it in a group of 1, you'll need instructor approval. Please try to find a partner: research can be lonely :)