Simulations

Synthetic Data Simulator for Horses

The project aimed to generate synthetic data focused on horse behaviors. This generated data can be utilized to train machine-learning models that can identify unusual behaviors in horses. The horse 3D model was rigged inside Maya, and a control rig was created within Unreal Engine for animation purposes. The second video demonstrates how the application operates. It can capture thousands of frames in mere seconds and produce the necessary data points. The captured frames are saved into a designated folder based on their session ID, while the generated data is exported in JSON format to an SQL database. For testing purposes, I have written an OpenCV script to visualize data.