The Edge computing board can reach 105 FPS and is able to perform over 1 trillion floating point operations per second as a dedicated neural network accelerator. The Myriad X has been designed to be able to process video and perform facial recognition in real-time. It is an ultra-low power VPU that requires very little energy. The package includes the Intel Vision Accelerator Design SW SDK, a supported framework consisting of TensorFlow, Caffe and MXNET, and Ubunto 16.04 and Windows 10.
For the development of applications that emulate human vision, RDS can also supply Intel’s Distribution of OpenVINO toolkit, which facilitates CNN-based deep learning inference at the edge. The toolkit extends workloads across Intel hardware, including accelerators and maximises performance. It speeds up time to market via a library of functions and pre-optimised kernels as well as including optimised calls for OpenCV and OpenVX.