Inside the deploy.prototext.txt file is architecture of model(layers), inside res10_300x300_ssd_iter_140000.caffemodel file are the weights for actual architecture model(layers).
face_detector_image.py file is for face recognition from images and face_detector_video.py file is for face detection from video.