CVGM: Computer Vision and Generative Models
Advisors:
Description:
This team will investigate methods of generative artificial intelligence to produce images or videos based on inputs that may include text, image, video, or music. This team will define metrics that evaluate generative models' abilities and limitations to produce desired outputs. These metrics may include (but are not limited) to (1) Can the models correctly produce outputs based on the inputs? (2) Can the models produce outputs quickly and provide interactions with users? (3) What are the strategies for the inputs in order to produce desired outputs? (4) Can the models generate diverse outputs and maintain consistent style?
This project can count as senior design.
Qualifications/Requirements:
The CVGM team can accommodate students with different skill levels. Beginners can help curate the data and survey literature, while students with more mathematical or programming skills can create new machine learning methods. Students interested in this topic are encouraged to talk to the advisor.
Relevant Skills
- Computer programming
- Computer vision
- Machine learning