Presenter(s)

Shreenivasan Duraivelan

Download

Download Project (139 KB)

Description

Trajectory Prediction is the problem of predicting the short-term (1-3 seconds) and long-term (3-5 seconds) spatial coordinates of various agents such as cars, buses, pedestrians. In Trajectory, Prediction we use the moment of an agent’s past trajectories to predict action or course the agent is going to make in the future. In our paper, we are using trajectory prediction to predict the future location and the movement of the players and ball in a basketball game. There are several Pedestrian trajectory prediction modules which have yielded great results. But applying the same to a basketball game will not yield great results since the pedestrian trajectory modules are generalized for the huge population. We propose a model which will learn the play style and the players position/roll, which will lead to better predictions in a game. We use Transformer network to implement the trajectory prediction algorithm along with a player behavior module which will increase the accuracy of the prediction.

Publication Date

4-22-2021

Project Designation

Graduate Research

Primary Advisor

Van Tam Nguyen

Primary Advisor's Department

Computer Science

Keywords

Stander Symposium project, College of Arts and Sciences

Group Trajectory Analysis in Sport Videos

Share

COinS