Computer Vision

The lessons are not mandatory

HOWEVER

We kindly suggest that you attend them for multiple reasons.

• You really learn something
• You will be rewarded during the final examination
• The topic you see in the classroom could become really interesting to you. It means that you could like to collaborate with the Visionlab or ask for a thesis with us

The exam consists of two tasks to perform:

• Present a paper (presentation with slides).
• Present a project related to the topic of the paper.

Maximum time per discussion: 15 minutes (e.g., 5 min for the paper + 10 min for the project; the group can decide how to manage the time).

The exam can be prepared in a group (maximum of 4 students).

The subjects can be assigned in two ways:

• By direct request: you know what kind of topic you would like to explore, and you can propose something related to a possible problem in the chosen field.
• By indirect request: you don’t have an idea of which topic you like to explore, so you can ask for a suggestion about a possible problem to solve.

The project requires minimal coding skills. You don’t have to implement a system from the beginning; you can exploit already-made code. However, you have to fit it into the topic of the paper that you chose and the problem you want to solve.

AI usage is allowed, such as cloning Git repos and any other supportive tools. HOWEVER, it is mandatory to highlight the personal contribution to the work produced.

The code and the presentation must be delivered a couple of days before the discussion.

The final grade depends on:

• Course attendance: 5%
• Final presentation: 40%
• Final project: 50%
• 1 -10% (yes, more than 100%) for
  • Creative ideas
  • Paper submission
  • Task complexity

The grade is proportional to the effort.

The discussion dates are chosen with the attendees to select the most popular days and allow students to participate.