Inspiration
Our inspiration for this project was our own personal interests in playing multiplayer FPS video games. Ever since the beginning of the pandemic, we found that video games are able to keep us in contact with out friends without physically meeting them. From that point on, us and a lot of other friends have been playing together quite often, so naturally we wanted to improve at the games that we played.
What it does
Our project analyzes the aim of a player during their warm-up period to predict how well they will do in their game. It does this by using previous data inputted by the user. If the player is unhappy with the prediction, they can warm-up more until the model tells them that they are ready.
How we built it
We built our hack using python, tkinter, and firebase. Python was used for most of the functions, such as Machine Learning, taking in user input, linking to firebase etc. The Tkinter library was used within python to coordinate the user interface of the program. Firebase was used to make the app.
Challenges we ran into
We ran into challenges with the program logic, specifically tracking the past and current location of the targets and mouse.
Accomplishments that we're proud of
We're proud of getting the tkinter animation to work, as well as displaying the graphs in the environment.
What we learned
We learned a lot about Machine learning: how it works and how to make a model. We learned a lot about tkinter as well, especially how to animate images. We also enhanced our learning of firebase.
What's next for Aim Analyzer
Currently, Aim Analyzer uses 3 tasks to judge the aim of the player. However, there are many more tasks that can be added to test various skills, such as tracking or hitting moving objects. In addition, we can improve our prediction modeler. Currently, we only consider scores for the 3 tasks and take the average, but later on we want to weigh the 3 scores and 3 accuracies differently to make the most accurate model and predictor of KDA and aim performance as much as possible. Another aspect we want to add is to be able to change the sensitivity used in the application to model the sense of the player used in the game. We would also like to make this a 3D program in the future as well, with a crosshair at the center and changing viewpoints based on movement of the crosshair (so the cross-hair will always be centered on the screen). We also will make the program more aesthetically pleasing and user friendly.
Log in or sign up for Devpost to join the conversation.