Automatic Video Quality Monitoring

Technology #3861

Questions about this technology? Ask a Technology Manager

Download Printable PDF

Image Gallery
Categories
Researchers
Nikil Jayant
Faculty Inventor Profile
External Link (www.ece.gatech.edu)
Nitin Suresh
Inventor Profile
External Link (www.linkedin.com)
Managed By
Rene' Meadors
Marketing Associate (404)385-0434

Background: The typical processed video is subject to coding and/or network errors. Subjective (as experienced by humans) video quality algorithms are statistical measurements dealing with how a particular video sequence is perceived by a viewer and therefore, related to the field of Quality of Experience (QoE). Such measurements are considered to be a key input for a successful business operation between a user and a content provider. 

Technology: Inventors at Georgia Tech have developed a novel video quality evaluation algorithm that evaluates the quality of processed video involving the measurement of subjective scores (e.g. meantime between failures - MTBF) based on objective measurements.  In its general form, the algorithm compares the received video signal with an anchor signal (e.g. information from the original signal) using established signal-processing methods such as autocorrelation, Markov models, neural networks, etc. The invention enables a no-reference system for monitoring video quality, where no reference to the original (undistorted) video information is available.

Potential Commercial Applications: The invention directly addresses the requirements and measures usage by video content providers, for example, in the entertainment video industry. Such a non-interactive subjective assessment method is applicable for the assessment of video, audio, and audiovisual quality of internet videos and distribution quality of television in multiple environments. 
Moreover, the method can be used for several different purposes including, but not limited to, comparing the quality of multiple devices, comparing the performance of a device in multiple environments, and subjective assessment, where the quality impact of the device and the audiovisual material are interlaced.  Finally, the method enables deployment of video quality metrics in a distributed network.


Benefits and Advantages: 

  • No need for original video reference
  • Reliable and flexible implementation 
  • No significant resources (processing power and  bandwidth) required 
  • Relevant for audiovisual communications and entertainment video
  • Suitable for assessment of various devices and/or operational environments