Streamlit + Tensorflow Webrtc streaming + inference Run streamlit to produce frames: $ bash start.sh Consume frames and run inference: $ bash consume.sh References https://pypi.org/project/asyncio-redis/