PULS
Foto: Matthias Friel
In this project seminar we will build an interactive system that can serve as the "eyes" of the user, through which the user can investigate a virtual environment. More specifically, the goal is to use (and potentially improve) current deep learning-based models that can describe images, in the context of an interactive system that is controlled with verbal (written) commands.
The students will gain familiarity with current state-of-the-art models in Language and Vision (a field at the intersection between computer vision and natural language processing), as well as approaches to modelling dialogue. A particular feature will be the emphasis on actually deploying these models in an interactive setting.
The project will be group work, with the division of tasks to be determined in the first meetings. (The meetings will be remote, and the work has to be coordinated remotely as well.)
The first meeting will take place on April 15, 10-12am, via Zoom;
https://uni-potsdam.zoom.us/j/62000158597 (pass code: 70925317 )
© Copyright HISHochschul-Informations-System eG