Contents:
- I. Introduction
- II. Automatic Impairment Detectors
- III. Manual Verification
- IV. Workflow Integration
- V. Conclusion
- Acknowledgement
- References
This technical paper describes tools developed to assess the visual quality of audiovisual content. Detected defects include things such as video breakup, noise/grain and sharpness issues, contrast/image and dynamic/brightness problems and blocking artifacts; they do not assess technical property conformance such as file integrity, format compliancy (as tools for those checks already exist). Section one identifies impairment detection algorithm, tool and system requirements that should enable the tools to work in both software and file based environments. Section two goes on to describe how the algorithms were developed. Section three presents three approaches to viewing and verifying tool results: from the Heinrich Hertz Institut, Joanneum Research Center and the RAI (Radiotelevisione Italiana). Section four describes how the developers are working to ensure the tools can be integrated into a SOA environment, especially focusing on the recent standardization efforts by the EBU and AMWA’s Joint Task Force on a Framework for Interoperability of Media Services in TV Production (FIMS).
This paper offers hope in the development of semi-automatic tools to assess the visual quality of audiovisual content, tools currently missing in the field and in high demand. It is best read by technical staff responsible for quality analysis who have an engineering background.