Intel’s new OpenVINO (Open Visual Inference & Neural network Optimization) toolkit, announced via press release Wednesday, could make it easier for developers to add artificial intelligence (AI) capabilities to Internet of Things (IoT) deployments at the edge.

Specifically, OpenVINO streamlines the process for adding deep learning inference and computer vision capabilities at the edge, the release noted. This includes a set of APIs that can be used to accelerate deployment of such capabilities.

The toolkit aims to address challenges including bandwidth, storage, latency, optimization, and scalability. As an end-to-end solution, OpenVINO might offer a lower-cost option than building out a similar set up piece-by-piece.

SEE: IT leader’s guide to the future of artificial intelligence (Tech Pro Research)

OpenVINO, and other Intel Vision products for that matter, can be used to port computer vision and deep learning inference from frameworks including “TensorFlow, MXNet, and Caffe, to Intel processor and accelerator technologies, including Intel CPUs, Intel integrated graphics, Intel field programmable gate arrays (FPGAs), and Intel Movidius vision processing units (VPUs),” the release said.

This breadth of options could make OpenVINO useful across a wide range of verticals, such as retail security systems or point-of-sale, as noted by Stephanie Condon of our sister site ZDNet.

Computer vision is a growing market, and MarketsandMarkets reported that it is expected to reach $17.38 billion by 2023. Additionally, Tractica data presented in the release predicts that deep learning revenue will hit $35 billion by 2025.

Developers can download the OpenVINO toolkit for free here.

The big takeaways for tech leaders:

  • Intel’s OpenVINO toolkit could fast-track the deployment of computer vision and deep learning capabilities at the edge, bolstering IoT efforts.
  • Solutions like the OpenVINO toolkit could help industries like retail in broader digital transformation efforts.