Our Blog
Research >> Efficient Compression and Transport of Plenoptic Video Content
Description
Current image and video compression standards such as JPEG 2000 and HEVC (High Efficiency Video Coding) are not specifically designed to take into consideration the intricacies of LF content, thus falling short of
efficiently compressing LF content and delivering the best possible viewer QoE. Some proposals involve use of HEVC, random codes,
and predictive coding for sub-aperture views. To address the compression challenges that LF introduces, JPEG has recently formed JPEG Pleno,
a framework that aims at defining new tools for improved compression of LF images. This, however, is a work in progress. In October 2017, a new Joint Video Experts Team (JVET) was created to develop a new video
compression standard with capabilities beyond HEVC, which has the opportunity to address the challenges of LF compression.
We are developing compression algorithms optimized for LF content. Working closely with our partner TELUS, we will ensure that our algorithms support compatibility with different networks and a range of playback devices.
Our schemes will be either integrated into the emerging JVET and JPEG Pleno standards or they will stand outside existing standards, paving the way for new standardization efforts. Packet loss caused by congestion and
transmission errors leads to a drop in the user’s QoE through freezing and rebuffering. The effect on quality is much more evident in the case of Multiview video due to hierarchical coding structures and multiple levels
of dependent layers, with error concealment more challenging for MVD than traditional 2D videos. As LF involves significantly more image/video data, it is expected that these issues will be more challenging.
We will design new rate adaptation methods and packet prioritization schemes for LF content as well as a new error concealment strategy.
Researchers
Selected Publications:
- N. Mehajabin, S. R. Luo, H. Wei Yu, J. Khoury, J. Kaur and M. T. Pourazad, An Efficient Random Access Light Field Video Compression Utilizing Diagonal Inter-View Prediction, 2019 IEEE International Conference on Image Processing (ICIP), Taipei, Taiwan, 2019, pp. 3567-3570.
- J. Khoury, M. T. Pourazad and P. Nasiopoulos, "A New Prediction Structure for Efficient MV-HEVC based Light Field Video Compression," 2019 International Conference on Computing, Networking and Communications (ICNC), Honolulu, HI, USA, 2019, pp. 588-591.