Undergrad Research Project - Video Understanding

Fall 2017

Xiang Xu
Kris Kitani
Project description

In the past few years, there have been a lot of interest in adversarial learning and generative models. Recently, GANs and other models have been applied to video generation or future frames prediction. The main topic for this project will focused on using generative models to learn and model complex interactions in videos. We will also borrow ideas from the field of reinforcement learning. For this project, I will work under Professor Kitani and we will have weekly meeting. The goal of this project is to create a new approach which incorporates both adversarial learning and RL and show that it surpasses existing methods.

Return to project list