Multiprocessing and threading of a facerecognition system

I have developed a face recognition application on python for surveillance cameras, i want the system to work in real time so i want to apply multiprocessing and threading: 1. Set up a separate thread to read frames from the camera: This thread will continuously read frames from the camera and put them into a queue. 2. Set up a pool of processes: Create a pool of processes (one for each CPU core) that will process frames from the queue. 3. Modify the pipeline to use multiprocessing: The pipeline should be modified so that each process takes a frame from the queue, processes it, and returns the annotated frame to the main process. Each process should run the entire pipeline for a single frame. 4. Annotated frames: The main process should display the annotated frames from each process in the correct way to assure a smooth live real time output