The goal of this project is to train an object detection algorithm on images generated from Unreal Engine. The Unreal Engine is a powerful tool for creating realistic 3D environments, and it can be used to generate large datasets of images for training machine learning models. In this project, the UnrealCV plugin is used to capture images from Unreal Engine and train an object detection algorithm on these images. Meta's Detectron2 library is used to train a Faster R-CNN model.