Build a Network Video Recorder

To ensure the highest standard of security, smart video surveillance systems need to be able to accommodate high-resolution video management, storage, and viewing. This solution demonstrates how to implement and take advantage of Intel® hardware platforms for video decoding, encoding, and optimization using various media stacks.

Target Operating System Ubuntu* 16.04 LTS
Time to Complete 45 minutes
Software Used

Intel® Distribution of OpenVINO™ Toolkit

Intel® Media SDK

Model Used person-detection-retail-0013

GitHub* (Python*)


What You Will Learn

This application uses the Intel® Media SDK for encoding and decoding high-resolution video streams into a server machine for storage and management.

Gain insight into the following solutions:

  • High-resolution video processing
  • Computer vision applications for IoT
  • Surveillance market IoT

Learn to build and run an application with these capabilities:

Manage high-resolution video.


Transmit video into a computer vision application for people detection.


Convert video to multiple formats to fit multiple standards.


How It Works

This application uses two video producing terminals in which their streams are decoded into a server machine.

Using GStreamer and the Intel Media SDK, the application captures video streams and encodes them into a format that can be accommodated by the server machine.


The stored video in the server machine is streamed into a computer vision application, which performs inference to detect people.


The output from the inference pipeline can then be passed to other applications for further processing.