Traditional Computer Vision

Accelerate and optimize low-level, image-processing capabilities using the OpenCV library and OpenVX* API.

Traditional computer vision remains an important approach to building and optimizing applications and solutions that have low processing power or are focused on smaller data sets.

The OpenVINO™ toolkit delivers access to OpenCV and OpenVX vision functions—complementary libraries that provide access to software algorithms and accelerate capabilities of CPUs and GPUs from Intel.

OpenCV: Optimized Functions for Intel® Accelerators

This library is typically used for processing general computer vision-related flows—such as real-time image and video—and for providing analytics and machine learning capabilities. Its large number of primitives are easy to customize, giving you a reasonable performance baseline.

Based on your critical path, achieve additional performance by using OpenVX or the Intel® Deep Learning Deployment Toolkit.

For more information, see the OpenCV Team site.

OpenVX*: A New Heterogeneous Computing Standard

OpenVX is an API that provides excellent image-to-image processing. Offering a higher level of abstraction, this standard API:

  • Facilitates portable, cross-platform performance by expressing workloads as connected data flow graphs
  • Allows you to interactively create these graphs
  • Optimizes the graphs by executing different workflow stages on the best nodes in the heterogeneous platform

For more information, see OpenVX from Khronos* Group.

Code Samples

Discover More Capabilities

Access traditional computer vision samples to use with your project.

Deep Learning for Computer Vision

Accelerate and deploy convolutional neural networks (CNN) on Intel® platforms with the Intel Deep Learning Deployment Toolkit (included).

Hardware Acceleration

Choose Intel®-based accelerators (CPUs, GPUs, FPGAs, and VPUs) that best harness your product performance.