Generation Codec Research Articles

Flow incorporated neural network based lightweight video compression architecture

The sudden surge in the video transmission over internet motivated the exploration of more promising and potent video compression architectures. Though the frame prediction based hand designed techniques are performing well and widely used but the recent deep learning based researches in this domain provided further directions of pure deep learning based next generation codecs. As the bandwidth over the internet is varying, adaptive bit rate representation is more suitable for video quality adjustment in tune with bandwidth variation. The proposed architecture comprises of end to end trainable video compression network consisting of majorly three modules namely-motion extension network, flow autoencoder and frame autoencoder. Frame autoencoder generates the individual compressed frames, flow autoencoder is used for optical flow based motion compensation chore and next frame is predicted by the motion extension network. The network is designed and evaluated in incremental manner. The analysis of the outcomes demonstrates the promising performance of the network quantitatively and qualitatively. Moreover, the results reveal that inclusion of optical flow based motion compensation network to the MotionNet architecture has enhanced the performance.

Object-Based Audio for Sports TV Production

Object-based audio (OBA) is the next exciting breakthrough in television production. It will provide personalization and an enhanced listening experience as revolutionary as sound was to motion pictures. The age of personalization has arrived, and TV consumers can view any program, at any time, and on virtually any media device. The next generation of audio encoders will have the ability to create OBA in television production and post-production. The beneficial features of OBA include audio personalization for language selection, dialogue enhancements, and options for the hearing impaired. OBA provides the viewer with the ability to customize their viewing for any type of program in any viewing setting. The future audio codec technologies will enable audio production mixers, producers, and broadcasters to produce customized audio for the viewer. This process begins at the original mix location and continues through the broadcast chain to delivery on any consumer device. These encoders will have the ability to emit surround sound and immersive sound with more than 100 channels with objects either separately or in combination with each other. Scene-based audio will also be a feature of the next generation codecs, enabling the mixer to represent the sound image instead of channels. This paper describes the evolution from current channel-based TV production to the next generation of multi-featured audio encoders with OBA and the potential benefits they will offer to all types of TV viewers.

Introduction to the Issue on Emerging Technologies for Video Compression

The ten papers in this special issue focus on emerging technologies for video compression. There are contributions related to four key research areas: 1) the merging HEVC standardization process; 2)advances in 3-D coding; 3) quality assessment for next generation codecs; and 4) new compression methods exploiting spatio-temporal content analysis and synthesis.

Subjective Quality Assessment of the H.264/AVC In-Loop De-Blocking Filter

Next generation codecs, benchmarked by the H.264/AVC standard, are providing substantial compression efficiency for the coding and transmission of video. Coupled with technologies offering larger transmission bandwidths over DSL, wireless and satellite networks, the capability of delivering high quality video services to the home is now a reality. The perceptual quality of the content delivered over communications networks will be crucial in ensuring a first-class customer experience. It is therefore important to assess the advantages and disadvantages of the optional features offered by next generation codecs. This paper describes a subjective assessment that was carried out to investigate the perceptual effects of switching the in loop de-blocking filter within the H.264/ AVC CODEC on or off. Although the filter is believed to substantially improve the perceptual quality of video, it has been suggested that in some cases negative perceptual effects can be produced. The H.264/AVC architecture allows de-blocking to be switched off in cases where there are limited processing resources or it is considered a negative perceptual effect may be introduced. This paper describes a study that examined the perceptual effects of de-blocking by employing a standardised subjective assessment methodology. The Absolute Category Rating (ACR) method was used to capture Difference Mean Opinion Scores (DMOS) for a range of video. Content was selected to span a wide and representative range of coding complexity. This content was then encoded at a variety of bit-rates to represent high, medium and low qualities. Results were used to examine the end-user perception of video quality when the de-blocking filter is switched on or off. The experimental design allowed the overall effects of the de-blocking filter to be examined and additionally the relationship between content and quality on the filter performance. The experiment found that the performance of the de-blocking filter was content-dependent. Results were used to discuss the advantages and disadvantages of in-loop de-blocking and there is an examination of content properties (e.g. spatial and temporal complexity) that influence the performance of de-blocking.