Adaptation methods for scalable graphics and animation on constrained devices

拟议的研究计划需要基于动态变化的上下文信息来适应可伸缩的多媒体内容。所讨论的可伸缩媒体是能够在空间上、时间上和在其信噪比方面缩小的单个“高质量”媒体文件(即，至少包含其自身的最佳版本的文件；例如，1080p视频)。这种缩放可以通过增量方案和/或存储冗余信息来实现；然而，最终目标是获得包含非常高的可伸缩性的高度压缩的文件(尽管这可能不是最终可以实现的)。上下文信息依赖于应用，但通常包括一系列信息，包括终端设备(例如移动电话)的能力和状况、网络连接、甚至用户的限制和能力。为了向用户提供适当的服务，上下文通常被认为是动态输入，根据环境、网络条件、在终端设备上执行的其他应用等而改变。适配过程通常通过基于所提供的上下文信息做出模糊判决(与直接二元判决相反)来将可伸缩媒体和上下文连接在一起，并且相应地正确地缩放媒体(并且希望动态地与上下文改变相关)。

The proposed program of research entails the adaptation of scalable multimedia content based on dynamically variable contextual information. The scalable media in question is a single "high-quality" (i.e. a file which contains at least the best version of itself; e.g. 1080p video) media file that is able to scale down spatially, temporally, and in terms of its signal-to-noise ratio. This scaling can be achieved through an incremental scheme and/or the storage of redundant information; however the ultimate goal is to obtain a highly compressed file containing a very high degree of scalability (although this may not be ultimately achievable). Contextual information is application dependent, but generally comprises a range of information including the capabilities and condition of the terminal device (such as a mobile phone), the network connection, and even the user's restrictions and capabilities. In order to provide a proper service to a user the context is generally considered as a dynamic input, changing according to the environment, network conditions, other applications being executed on the terminal device, etc. The adaptation process essentially connects scalable media and context together often by making a fuzzy decision (as opposed to a direct binary decision) based on the contextual information provided and correctly scaling the media accordingly (and hopefully dynamically relating to the contextual changes).