Methods to Simplify Object Tracking in Video Data

TL;DR

This paper introduces simple, color-based algorithms for object tracking in videos, integrated into a free browser-based tool, improving robustness and accessibility over complex image processing methods.

Contribution

It presents novel, straightforward algorithms that outperform traditional methods by focusing on pixel color analysis, implemented in an accessible, open-source software.

Findings

Algorithms are more effective than complex image processing methods.

The tool is browser-based and platform-independent.

Improves object tracking reliability for educational use.

Abstract

Recent years have seen an explosion of interest in analyzing the motion of objects in video data as a way for students to connect the concepts of physics to something tangible like a video recording of an experiment. A variety of software exists for students to look at individual frames and click on the object to infer the x,y position. Some of these tools include a capability to automatically identify the position of the object in the frame. But it is not unusual, especially when inexperienced users are recording the video and configuring the program, for these algorithms to struggle to "lock on" to the moving object. In this paper, we both include some general advice to help object tracking algorithms locate an object and we provide our own algorithms that are simpler and potentially more effective than the sophisticated image processing algorithms that are currently being used. These algorithms focus not on a "template image" of the moving object but instead distinguish between the object and the background by analyzing only the colors of individual pixels. These algorithms are built into a free and open source program called the STEMcoding Object Tracker (http://go.osu.edu/objecttracker) which works in the browser (without any downloads) and is compatible with a variety of operating systems including chromebooks.