New codecs to tame the video data flood
At IBC 2019, Fraunhofer HHI researchers will present the latest version of the Versatile Video Coding (VVC) reference code and the first prototype of a live software decoder for the VVC codec. In addition, Fraunhofer HHI and Volucap GmbH will present a novel capture studio and a processing workflow for high-quality volumetric video productions for future VR/AR media productions.
Compressed video data is growing faster than ever before. Already today, they make up by far the highest proportion of bits on the Internet and in mobile data traffic. This illustrates the need for more efficient compression beyond the current High Efficiency Video Coding Standard (HEVC). In order to meet this demanding challenge, the ITU-T Video Coding Expert Group (VCEG) and the ISO/IEC Moving Pictures Expert Group (MPEG) have already started working together in the Joint Video Experts Team (JVET). After leading technology companies (as well as Fraunhofer HHI) have submitted promising proposals for HEVC succession, the standardization of the Versatile Video Coding (VVC) called HEVC successor began in April. Standardization is expected to be completed by 2020 and the final standard will achieve 50% bit rate reduction compared to HEVC.
At the show, Fraunhofer HHI will present the latest version of the VVC reference codec to the public. This version shows significant bit rate reduction compared to HEVC for a wide range of video content from High Definition (HD) to High Dynamic Range Ultra-HD. In addition, Fraunhofer HHI presents the first prototype of a software live decoder for the VVC codec.
The Fraunhofer Heinrich Hertz Institute and VoluCap GmbH present a novel and innovative capture studio as well as a processing workflow for high-quality volumetric video productions for future VR/AR media productions.
Volumetric video production is a new technology that creates three-dimensional virtual spaces instead of two-dimensional images. The technique is considered to be the next “big thing” in video production. The scenery to be depicted is not recorded by one or two cameras. Instead, a large number of cameras film the scene from just as many different angles. Moving point clouds are calculated from the recording data (up to 600 gigabytes per minute). These point clouds result in a virtual 3D world. In post-production – this is the basic idea – users can move around freely in the filmed environment and take almost any perspective for the final video.
In June 2018, the first volumetric video studio on the European continent was opened at Filmpark Potsdam-Babelsberg. Real people are recorded with several high-resolution cameras in a professional studio environment. A powerful processing suite automatically generates dynamic 3D models that can be integrated into AR/VR applications. The system supports diffuse or synchronized scenic light from any direction, automatic keying and flexible multi-camera arrangements.
More information: https://www.hhi.fraunhofer.de/en.html
Related articles:
Holographic-like video conferencing: a killer app for AR?
Making volumetric displays out of glycerin-filled tanks
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
