Release 2019 R3

This release contains GStreamer* Video Analytics plugins that bring Deep Learning Inference capabilities to open-source framework GStreamer* and helps developers to build highly efficient and scalable video analytics applications. The solution:

• Extracts insights from video stream(s) using object detection, classification and recognition CNN models and sends video metadata to application or cloud service for further processing  
• Leverages hardware acceleration for media and inference operations and heterogeneous execution across Intel CPU and GPU
• Provides flexible mechanism to quickly construct video analytics pipeline from highly optimized building blocks  
• Deployable on both edge devices and cloud infrastructure, including deployment in Docker containers 
• Can be used for various applications such as video surveillance and security, smart city, retail analytics, ad insertion and others 

New features in this release:

• Support of VCAC-A (Visual Cloud Accelerator Card – Analytics) with all VCAC-A specific performance optimizations enabled in default Docker build
• Introduced pre-processing on Intel® Graphics via VAAPI, significantly improving video analytics performance on the systems with Intel's integrated and discrete Graphics
• Improved realtime stream density of video analytics workloads on Intel® Xeon® Scalable Processors
• New GStreamer element for convenient FPS measurements including multichannel cases (gvafpscount) and accompanying sample
• Updated Docker* base image to Ubuntu* 18.04 and revised dependencies list

Optimization notice:

For optimal performance on Intel® Xeon® Scalable Processors, use "pre-proc=ie" option in inference elements. Please note that this is currently supported only with open source version of OpenVINO™ Toolkit.

Release details including known issues and system requirements are in available in the attached release notes.
Getting started is available on the Wiki.