Do you need to detect when a person is in a room but, for whatever reason, can't track them with a camera? This project aims to use a handful of simple sensors (temperature, CO2, light, and humidity) and some neural nets (from TensorFlow's framework) to predict the presence of a human without such a camera.
I'm not sure when you would have the processing power to run a neural net but not the processing power to run a camera peripheral but this project seemed like a good way to learn more about TensorFlow.
This project relies heavily on data gathered by Luis M. Candanedo and Véronique Feldheim. Feel free to check out their paper to read more about the original project:
Accurate occupancy detection of an office room from light, temperature, humidity and CO2 measurements using statistical learning models. Luis M. Candanedo, Véronique Feldheim. Energy and Buildings. Volume 112, 15 January 2016, Pages 28-39.