Research Article
Vision-Based Car Control System for Gaming with Integrated Driver Performance Rating
@INPROCEEDINGS{10.4108/eai.19-12-2025.2362630, author={Govind Gonnade and Akhil Gupta and Shadab Perwez and Nitin Rakesh and Monali Gulhane and Firdous Sadaf Mohammad Ismail6}, title={Vision-Based Car Control System for Gaming with Integrated Driver Performance Rating}, proceedings={Proceedings of the 1st International Conference on Computational Techniques in Data Science, IICTDS 2025, 19-20 December 2025, Chandigarh, India}, publisher={EAI}, proceedings_a={IICTDS}, year={2026}, month={11}, keywords={Vision-based control Gesture recognition Gaming interface Driver performance evaluation Real-time hand tracking}, doi={10.4108/eai.19-12-2025.2362630} }- Govind Gonnade
Akhil Gupta
Shadab Perwez
Nitin Rakesh
Monali Gulhane
Firdous Sadaf Mohammad Ismail6
Year: 2026
Vision-Based Car Control System for Gaming with Integrated Driver Performance Rating
IICTDS
EAI
DOI: 10.4108/eai.19-12-2025.2362630
Abstract
This paper introduces a car control system reliant on vision that uses real-time hand gesture recognition as a formal gaming application. The system is based on OpenCV and MediaPipe, coded in Python, which allows the virtual vehicle to be controlled without any physical input devices and thus creates an immersive, low-latency gaming experience. This also leads to the creation of a novel driver performance rating which is based on a heuristic, i.e. technically unrelated, metrics such as steering precision, reaction time, and control consistency (task-based), without any reliance on internal game physics or telemetry data. The gesture component is based on the landmark detection algorithms, which take a standard laptop camera as input and translate it into game controls. The system has been tested with 12 participants, confirming its usability and the capability to deliver actionable feedback that differentiates between driving skill levels while still being compatible with various hardware configurations.