NOISE APPLICATION AGAINST FPS AIMBOT VIA DEEP LEARNING ALGORITHMS
by Jaeyun Hwang
Category: Computer Science
Abstract – The use of aimbots in first-person shooter video games is a common cheating method. Researchers experimented with using noise as a way to disrupt character detection by machine learning aimbots. Human Action Detection was used to simulate a program similar to what aimbots may use. Various types of noise were used, resulting in a 10% reduction in detection accuracy. An experiment was done for the game Overwatch, with noise applied to reduce detection accuracy. While the result was clear, the real-life application for noise in games is uncertain. Further research is required to improve noise application while reducing visual impact.