This is a demo show the example of the face recognition. Once it detects faces, the frame rate drop to 0.1FPS. If using the CPU to process landmarks, there seems no problem.
Is the landmarks-regression-retail-0009.xml model not compatible with Movidius?
Thank you in advance.
#Go to face recognition example cd /opt/intel/openvino/deployment_tools/open_model_zoo/demos/python_demos/face_recognition_demo #Download models python /opt/intel/openvino/deployment_tools/open_model_zoo/tools/downloader/downloader.py --list models.lst -o . #Make face DIR mkdir faces #Run python script python ./face_recognition_demo.py -m_fd intel/face-detection-retail-0005/FP16/face-detection-retail-0005.xml -m_lm intel/landmarks-regression-retail-0009/FP16/landmarks-regression-retail-0009.xml -m_reid intel/face-reidentification-retail-0095/FP16/face-reidentification-retail-0095.xml -fg faces -d_fd HDDL -d_lm HDDL -d_reid HDDL
Thanks for reaching out.
We ran your command on both an Intel® Neural Compute Stick 2 with MYRIAD plugin and a Mustang card with 8 VPUS with HDDL plugin and they both worked. Could you please tell us the specific Movidius board you are using?