I wanted to start developing a project with image recognition. I want to know the difference between the Intel Movidius Neural Compute Stick and TensorFlow to develop this project Any help would greatly be appreciated. What is the difference between them and what is better for accuracy? Thanks, Aditya
The Intel Movidius Stick is an USB stick for running a neural networks (that are programmed in Tensor Flow) on resource scare devices such as a Raspberry Pi (in order to reduce the traffic by sending data to cloud service or a server with more capacity where your neural network runs).
Tensor Flow is a programming framework which can be used as python library for the implementation of neural networks.
You are comparing an execution environment with an implementation language. The accuracy should not depend on them, instead I would considering on your model architecture (type of cnn) and your training as the most influential factors on accuracy of your test data.