Rate-distortion optimized SVC bitstream extraction for heterogeneous devices : A preliminary investigation

Abstract

The emerging SVC standard enables partial extraction and decoding of video bitstreams, and thus allows various viewing devices to adapt their video reception and playback according to devices capability and network performance. This desirable feature, however, comes with a caveat: the parts of a bitstream needed for providing good quality playback at different devices may differ significantly depending on the visual characteristics of video programs, the quantization and dependency settings of SVC encoders as well as the display formats of viewing devices. In this paper, we present the results of our preliminary investigation on the intricate relations among these factors. We discovered a set of constraints on the setting of quantization parameters and the choices of reference layers for inter-layer dependencies that ensures good rate-distortion trade-offs and regular bitstream; extraction paths. We called these constraints, the adaptation rules for inter-layer encoding, and their resulting outputs, the well-adapted layers/bitstreams. We further discovered that bitstream extraction by different viewing devices follow predictable paths if the SVC bitstream is well-adapted and the playback process extracts a complete set of interdependent network abstraction layer (NAL) units in every refinement step. These regular extraction paths enabled us to develop an efficient bitstream extraction algorithm based on local optimization of rate-distortion ratios. The experimental results using standard testing sequences confirmed our findings.