Report an accessibility problem

Engineering | Undergraduate Research

Predicting Video Steams using Deep Learning

, ,

Dr. Heni Ben Amor is an assistant professor in the School of Computing, Informatics, and Decision Systems Engineering (CIDSE). His lab, the Interactive Robotics Lab, is currently looking for volunteer or FURI students to aid in one of his research projects.

In this project we will use Deep Learning techniques in order to process video streams. The goal is to predict a video stream 1 second into the future using neural networks. To this end, we will use an existing implementation for deep learning from video streams. The task of the student will be to perform various experiments in which the movement of a falling, bouncing ball will be predicted from several video frames. The implementation will also be extended in order to allow a robot to use the generated predictions to catch it later on. This project requires knowledge of Python and general programming skills. The student should expect to work approximately 6 hours a week on this project.

Interested students should contact Dr. Ben Amor at



Posted on

September 19, 2017