Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.694452
Title: High Efficiency Video Coding (HEVC) tools for next generation video content
Author: Wang, Yanxiang
Awarding Body: University of Sheffield
Current Institution: University of Sheffield
Date of Award: 2016
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
The emergence of high resolution video content together with an increasingly broad range of content types demands highly efficient solutions for video coding. The new types of video content are defined as next generation content. High Efficiency Video Coding (HEVC) is emerging as an effective video compression technique that has been developed for new formats and new types of video content. This thesis aims to develop coding tools in HEVC for two types of next generation video content: screen content and Ultra Highdefinition (UHD) content, to improve the coding efficiency or the overall visual quality. For screen content, several colour formats are discussed and evaluated by comparing against natural content. The purpose is to convert the input RGB 4:4:4 video sequence into a luma-chroma model and maximise the de-correlation of the colour components. On the basis of the obtained results, a new design of the lifting-based Predict-and-Update colour conversion scheme is proposed for screen content. One of the proposed method achieved PSNR BD-rate saving up to 8.86%. Contouring is one of the artefacts in HEVC compressed UHD content. To reduce the contouring artefacts in UHD content, there are three methods proposed in this thesis. Multi-scale dithering is a contouring removal method applied as a post-processing method. Pixel mapping is another algorithm designed for contouring artefacts removal. It can be implemented as either a pre-codec or an in-codec approach. Both objective and subjective quality metrics are used to evaluate the performance of the proposed coding solutions. The results show that the proposed dithering method achieved the MOS BD-rate saving up to 85%, and one of the proposed pixel mapping method achieved the MOS BD-rate saving from 23% to 78%.
Supervisor: Charith, Abhayaratne ; Marta, Mrak Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.694452  DOI: Not available
Share: