Video Prediction is a computer vision task that involves predicting future frames given past video frames (see https://paperswithcode.com/task/video-prediction for some existing papers and benchmarks). It can be used for a variety of applications, including autonomous vehicles, robotics, and weather forecasting. The goal of this project is to develop a deep learning based video prediction method that can achieve sufficient efficiency and accuracy. The student should be familiar with Python programming. Experience in deep learning and computer vision will be beneficial.