This is easily done using opencv's cmd line tool opencv_annotation. In trainingdatageneration.py, it takes screenshots of a certain part of your screen, converts them to grayscale and outputs them into either a positive or negative folder, depending on the button pressed (which can be done while playing CSGO normally, no need to tab out).Īfter getting screenshots and trying for a 2:1 ratio of positive to negative images, which I found to be the optimal choice online, you need to annotate the positive files. To quit the aimbot, press the QUIT_KEY key (default is k).Ī Haar Cascade Classifier is an object detection method that trains on "positive" and "negative" images and creates a model that takes an image and outputs an array of rectangles which represents where it thinks a certain type of object exists (which is defined by what you choose to the desired object in the positive images). You play normally as you would in CS:GO, but when you want to snap to a player, click the key that you set in the AIMBOT_CHARACTER constant (default is l). This aimbot only works well against Counter Terrorists (as their color contrasts better against backgrounds than Terrorists). This is my attempt of making a CSGO Aim assist tool using Python, OpenCV and Haar Cascade Classifiers.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |