Search Papers | Poster Sessions | All Posters
Poster B7 in Poster Session B - Thursday, August 8, 2024, 1:30 – 3:30 pm, Johnson Ice Rink
A Novel Method for Evaluating Expert Multiple Object Tracking Using Competitive Esport
Trent Simmons1 (), Ashwini Khedkar, Nitika Jain, Aswin Lakshmanan Sriram, Shravan Dinakaran, Leanne Chukoskie; 1Northeastern University
This study investigates differences in gaze performance strategies for expert and novice eSports players. Using a novel methodological approach combining eye-tracking and computer vision object detection, we present evidence that a fast-paced Esport—Rocket League—can offer an appealing alternative to traditional multiple object tracking (MOT) tasks. Our approach is able to make gaze performance comparisons across different levels of expertise, including complex MOT gaze strategies like center-looking. Our preliminary results show that experts look significantly longer at game objects, and both groups use center-looking as their primary gaze strategy. We find the use of gaze analysis in Esport to be an exciting method to examine expert performance in dynamic and richly complex scenarios.
Keywords: Gaze Multiple Object Tracking Expertise