Intel Distribution of OpenVino toolkit

Intel Distribution of Open VINO toolkit-Develop Multi platform Computer Vision Solutions.

Product Availability: AGAINST PO

Country Of Origin : USA

FEATURES

  • Develop applications and solutions that emulate human vision with the Intel® Distribution of OpenVINO™ toolkit. Based on convolutional neural networks (CNN), the toolkit extends workloads across Intel® hardware (including accelerators) and maximizes performance.
  • Enables CNN-based deep learning inference at the edge
  • Supports heterogeneous execution across computer vision accelerators—CPU, GPU, Intel® Movidius™ Neural Compute Stick, and FPGA—using a common API
  • Speeds up time to market via a library of functions and preoptimized kernels
  • Includes optimized calls for OpenCV and OpenVX*.
What’s New in the 2018 R5 Release
  • Extends neural network support to include long short-term memory (LSTM) from ONNX*, TensorFlow*, MxNet frameworks, and 3D convolutional-based networks in a preview mode (CPU only) to support additional, new use cases beyond computer vision.
  • Introduces the Neural Network Builder API in a preview mode, which provides the flexibility to create a graph from simple API calls. Directly deploy it using the Inference Engine without loading intermediate representation (IR) files.
  • Delivers a significant boost in CPU performance, especially on multicore systems, through new parallelization techniques.
  • Provides INT8-based primitives for Intel® Advanced Vector Extensions-512, Intel® Advanced Vector Extensions 2, and Single Instruction Multiple Data (SIMD) extensions (SSE4.2) platforms that deliver optimized performance on Intel® Xeon®, Intel® Core™, and Intel® Atom processors.
  • Supports Raspberry Pi* hardware as a host for the Intel Movidius Neural Compute Stick 2 (preview). Offload your deep learning workloads seamlessly to this low-cost, low-power USB stick that’s based on the Intel® Movidius™ Myriad™ X technology. It also supports the previous generation.
  • Adds three optimized pretrained models (a total of 30 in the toolkit):
  • Text detection of indoor and outdoor scenes
  • Two single-image, super-resolution networks to enhance the resolution of an input image by a factor of three or four.