Abstract-Adaptive multimedia streaming aims at adjusting the transmitted content based on the available bandwidth such as losses that often severely affect the end-user perceived quality are minimised and consequently the transmission quality increases. Current solutions affect equally the whole viewing area of the multimedia frames, despite research showing that there are regions on which the viewers are more interested in than on others. This paper presents a novel region of interest-based adaptive scheme (ROIAS) for multimedia streaming that when performing transmission-related quality adjustments, selectively affects the quality of those regions of the image the viewers are the least interested in. As the quality of the regions the viewers are the most interested in will not change (or will involve little change), the proposed scheme provides higher overall end-user perceived quality than any of the existing adaptive solutions.Index Terms-multimedia streaming, content adaptation, user perceived quality, region of interest.
I. INTRODUCTIONsustained growth in the number of broadband connections to residential users and business premises is expected, as part of a trend towards all-IP networks [1]. They provide a low cost, high bandwidth infrastructure that enables the distribution of rich content services, many of them multimedia-based such as digital and interactive video and audio -including video on demand and voice over IP. The success of this trend relies heavily on the users' quality of experience and on the price they have to pay. In order to reduce this price, the network operators strive to ensure a high utilisation of their infrastructure and simultaneously serve many customers. However, this reduces the quality of their services and is at odds with the operators' wish to deliver high quality streamed multimedia to their customers. Therefore there is a need for a solution to stream multimedia data -with its high bandwidth and stringent timing requirements -such that it accommodates these constraints. A typical architecture for multimedia streaming over IP networks is presented in Figure 1.Adaptive solutions [2] were proposed for streaming G.-M. Muntean is with the
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.