Skip to content. | Skip to navigation

Personal tools
Log in

WG2: Critical Review and Insight into HDR Manipulation

Goal
To critically examine all aspects of manipulating the digital representation of real world lighting and then converting this to a video stream.

Objectives
To propose a file format for the most efficient representation of a stream of frames
which may have been manipulated and develop a metric for comparing the quality of the compressed manipulated frames. New compression algorithms are urgently required to facilitate the efficient exchange of HDR imagery across Europe without losing any desired quality.

Methodology
Captured real world lighting can be manipulated to support a wide range of applications. The most common applications are:

  • Post production: This is the many different processes which may be performed on recorded frames. Post production includes editing the footage, adding visual special effects, transfer of film to video, colour correction etc.
  • Image Based Lighting: This technique uses the captured HDR lighting as an area light source and environment map to accurately recreate the real world lighting conditions in a virtual scene. Virtual objects can be added to real world scenes and appear as if they were physically “there” in reality.
  • Scene analysis: Computer vision techniques can be used to analyse the footage to identify and track certain moving features, such as objects and faces to be used, for example in security and traffic control applications, or to enable a robot to successfully navigate an environment.

Implementation
The applications which WG2 will investigate are many and these applications often adopt very different approaches in how they work with the captured HDR images and video. Many of the applications are well established in the commercial sector, whereas others are just starting to emerge from research labs across Europe. In addition a number of related standards already exist. For example, ISO 26868:2009 establishes a data compression algorithm applicable to twodimensional digital spatial image data from payload instruments and specifies how this compressed data is formatted into segments to enable decompression at the receiving end.

WG2 will be very interdisciplinary and will take great care to ensure harmony between the different communities who form part of it, including researchers and practitioners in computer graphics and computer vision, film makers, professional photographers etc.

The critical review and insight of manipulation methods will investigate the methods themselves and the process which follows to enable the manipulated images to be passed on within the HDR pipeline to the delivery stage. This will include:

  • Identifying the cluster of methods which are used in each application area
  • Determining if any of these methods span more than one application area
  • Specifying the sectors engaged with each application area
  • Specifying current and emerging techniques for encoding manipulated HDR frames
  • Fully evaluating these encoding techniques in relation to their application area with regard to their quality, data storage efficiency, range of lighting, colour, spectral information that they can handle etc.

Expected Outcomes
WG2 will develop a systematically collected knowledge pool of existing HDR manipulation methods and encoding techniques, including:

  • Which methods are used for which applications and across which sectors
  • Who are the key stakeholders for these methods
  • The strengths and limitation of each encoding technique
  • The likely disruption that would arise if a technique were to be replaced or improved

This knowledge pool will provide the key insights into what is required from a comprehensive file format for manipulated HDR frames and an HDR video stream. The quality metric would allow a quantitative evaluation of all encoding schemes. Furthermore, knowledge of existing compression schemes will provide important insights into the development of the reference implementation of
the compression benchmark. This knowledge pool will be available to all other WGs and will be made publically accessible to the wider HDR community.

Deliverables
WG2-D1: Proposed standard file format for an HDR video stream
WG2-D2: Quality metric for comparing the quality of compressed HDR images and video
WG2-D3: A reference implementation of a benchmark HDR video compression algorithm against which future compression techniques can be evaluated.


INESC TEC - Laboratório Associado   European Partners