Autonomous decision making in layered and reconfigurable video coders

Layered compression and reconfigurable coding paradigms require coupled encoding decisions among layers or codec components in order to achieve optimal distortion performance under buffer constraints. Moreover, due to the dynamic and time-varying source characteristics, and temporal coupling of encoding decisions through the buffer constraints, both the immediate and long-term rate-distortion impact of encoding decisions must be considered. In many scenarios, however, it may be impractical to make such joint coding decisions. To address this problem, we propose a framework for autonomous decision making in layered and reconfigurable coders, which decouples the decision making processes of the various layers or codec components using a layered Markov decision process.