Is there a way to execute template matching algorithms over the Movidius VPU ? I see that Movidius is generally used for deep learning but I need to execute some template matching algorithms which executes painfully on RPi3.
Note: Deep learning is not a good option for my current situation.
Currently the Movidius NCS (Neural Compute Stick) is designed to work with Convolutional Neural Networks. but that may change in the future.
Intel Movidius stick enable rapid prototyping, validation, and deployment of deep neural network (DNN) inference applications at the edge. The low-power vision processing unit (VPU) architecture enables an entirely new segment of AI applications that are not reliant on a connection to the cloud.
Currently it supports two frameworks Caffe & Tensorflow.
Please find the link below to get more information on relese notes :
https://movidius.github.io/ncsdk/release_notes.html
Links mentioned below will give you more insights on Movidius ncsdk (It includes software tools, API and examples)
https://developer.movidius.com/start
https://movidius.github.io/ncsdk/install.html
For your kind information , RPi3 + Movidius NCS for CNN algorithms will give you good performance.