Hi everyone,
I would like to share our high-speed AI thermometer project with Jetson AGX Xavier.
The main idea of this project is to perform face detection directly from a thermal imaging camera rather than a regular RGB camera. The object detection model is SSD, and applied 8-bit quantization via TensorRT to improve the inference speed. Also I applied pipeline structure to divide the entire system into two stages to work in parallel.
I got a lot of inspiration and ideas from other projects, and I wrote them down as references in the github repo.
Have fun and feel free to give feedback. Thank you.
Github: GitHub - aiha-lab/AI-thermometer
