I read from this page on what we should do to compile the tensorflow models into graph form for use on the Movidius Stick- https://movidius.github.io/ncsdk/tensorflow.html. Just wondering if FasterRcnn ResNet 50 is supported as well because I am not sure what does:
If you are otherwise training your own TensorFlow network model for use with the NCSDK and Neural Compute API, follow the Guidance for Compiling TensorFlow Networks.
refer to exactly like must they be models that are part of the supported networks shown below?
Inception v1 Inception v2 Inception v3 Inception v4 Inception ResNet v2 MobileNet_v1_1.0 variants: MobileNet_v1_1.0_224 MobileNet_v1_1.0_192 MobileNet_v1_1.0_160 MobileNet_v1_1.0_128 MobileNet_v1_0.75_224 MobileNet_v1_0.75_192 MobileNet_v1_0.75_160 MobileNet_v1_0.75_128 MobileNet_v1_0.5_224 MobileNet_v1_0.5_192 MobileNet_v1_0.5_160 MobileNet_v1_0.5_128 MobileNet_v1_0.25_224 MobileNet_v1_0.25_192 MobileNet_v1_0.25_160 MobileNet_v1_0.25_128
There is an older thread about running ResNet on NCSDK2. Take a look!