Considerable progress has been made in the field of video coding [e.g., versatile video coding (VVC)] for the purpose of improving compression efficiency (e.g., reducing the amount of data needed to store or transmit a video). Previous generations of codecs have consistently managed to boost compression efficiency by a factor of roughly two. Just as relevant, however, is the progress that has been made in enabling video applications through design of the codec high-level syntax and the system integration. Some of these advances are already taken for granted (e.g., fast forwarding or streaming a video over the internet).
The Multimedia Communications group is exploring the limits of existing video codecs and working towards enabling new applications through developing technologies, contributing to standardization, and internal product work.1
Bitstream extraction and merging
There is an increasing demand for videos that convey 360-degree material (e.g., to capture live sports events). However, these high-resolution videos require immense amounts of decoder resources. To avoid this issue, we contribute to aspects of VVC, which allow the strategic enhancement or reduction of video resolution based on a user’s needs (i.e., via advanced picture partitioning methods). For instance, if a user has a certain viewing orientation, we reduce the resolution outside of the field of view, reducing the unnecessary consumption of resources. This method is demonstrated in the figure below.
Conformance of bitstreams and decoders
1 These activities take place within the framework of the Joint Video Experts Team, a collaborative effort by Video Coding Experts Group (Q6/16) and International Organization for Standardization (ISO)/International Electrotechnical Commission (IEC) Joint Technical Committee (JTC) 1 SC29/WG11 (Moving Picture Experts Group) in charge of developing the VVC.