deep-head-pose

deep-head-pose copied to clipboard

I'm trying to run this code I have no idea of how to replace the keyowrds in this command ' python code/test_on_video_dlib.py --snapshot PATH_OF_SNAPSHOT --face_model PATH_OF_DLIB_MODEL --video PATH_OF_VIDEO --output_string STRING_TO_APPEND_TO_OUTPUT --n_frames N_OF_FRAMES_TO_PROCESS --fps FPS_OF_SOURCE_VIDEO ' Can you give me a simply example or exlpain of each keyword

--snapshot is the path to hopenet weights file, downloadable at the provided links in the README.md. --face_model is the path to the dlib face detector weight, download the mmod_human_face_detector.dat from dlib repository and put the path here. --output_string string name that will be appended when saving the output as: output/video/output-%OUTPUT-STRING.avi --n_frames number of frames that need to be processed in the video. --fps framerate of the video.

where to find mmod_human_face_detector.dat? I tried but couldn't find it. Has it been removed?

Hi @khandriod , I think that's just the dlib face_detection model. You can download it from other source, just google it, at least that's how I've fixed it until now.

thanks, I have found it. I am struggling with to get results. One question I have, Do I have to train the model or I can make inference using per-trained model?