Car Detection over Network using Yolov8 in Forza Horizon 4

In the realm of autonomous vehicles, object detection holds a pivotal role in enabling accurate perception and safe navigation within complex environments. This study incorporates these domains by proposing a vehicle detection system employing YOLO v8 within the popular game Forza Horizon 4. YOLO v8's real-time, accurate object detection prowess aligns seamlessly with the game's dynamics, enriching the user experience through improved vehicle tracking. The research encompasses YOLO v8 model training on an expansive vehicle dataset, adaptation for Forza Horizon 4, and rigorous evaluation, comparing accuracy, speed, and efficiency against existing methods. The methodology entails dataset preparation, YOLO v8 configuration, object detection, result evaluation, optional fine-tuning, and iterative testing and optimization. By scrutinizing two data transmission protocols, RTMP and HTTPS, the research accentuates RTMP's superiority in real-time image data transmission for object detection, surpassing HTTPS in processing speed. While RTMP excels in low-latency scenarios, HTTPS augments security through encrypted data transmission. The findings prove RTMP's speed in image data transmission for real-time object detection, whereas HTTPS's secure communication advantages.