Project Team
Students
Love-Divine Onwulata
Computer Science
Abington
Faculty Mentors
Nate Wagenhoffer
Abington
Computer Science
Project
Project Video
Project Abstract
This study presents a novel approach to tracking the positional velocities of a flying disc using a combination of advanced computer vision techniques. Our methodology integrates optical flow algorithms with depth estimation methods to achieve accurate three-dimensional tracking of the disc’s trajectory and velocity. We employed optical flow models, including VideoFlow and RAFT, to capture the disc’s motion across consecutive video frames. These models provided precise pixel-level movement data, enabling us to calculate the disc’s velocity in the two-dimensional image plane. To extend our analysis to three dimensions, we incorporated the MiDaS depth estimation algorithm. This allowed us to infer the disc’s distance from the camera at each point in its flight path, providing crucial depth information for accurate velocity calculations. Additionally, we implemented object identification techniques to isolate and track the disc throughout the video sequence. This step was essential for distinguishing the disc from background elements and other moving objects in the scene. By combining these methods, we developed a robust system capable of tracking the flying disc’s position and velocity in three-dimensional space with high precision. Our approach overcomes limitations of traditional two-dimensional tracking methods and offers improved accuracy in velocity measurements, particularly for complex flight paths involving significant changes in depth. This research has potential applications in projectile motion analysis, and the development of autonomous systems for object tracking and interception.
Evaluate this Project
Use this form link to provide feedback to the presenters, and add your project evaluation for award(s) consideration.