Artificial Intelligence Image Recognition (AIIR) is a Python project that aims to develop a CNN model that can be used for image recognition. It was divided into two parts.
The first one aiming to expand its participants' knowledge about applications and usage of Artificial Intelligence in modern programming challanges, and to train them in using such solutions in actual real life problems. In this part two different datasets are used, to develop optimal data preparation techniques, find best network models and evaluate their results.
The second part is to use gained knowledge and developed tools in cooperation with another project: Project Webcam, which uses a digital camera to capture hand gestures in real time. Project AIIR will then be used in a basic classification problem, to determine what gestures were shown.
The script requires you to have a dataset unpacked in a directory with the following path: "datasets/gestures_dataset". It is possible to change that, which can be done in the first line of the main function.