Optimal Partitioning of Fine-Grained Scalable Video Streams

The increased popularity of video streaming over the Internet attracts numerous clients. These clients are quite heterogeneous in terms of network bandwidth and processing capacity. To accommodate this heterogeneity, fine-grained scalable (FGS) coding of video streams has been proposed in the literature. FGS streams are composed of two layers: base layer, which provides basic quality, and a single enhancement layer that adds incremental quality refinements proportional to the number of bits received. The base layer uses nonscalable coding which is is more efficient in terms of compression ratio than scalable coding used in the enhancement layer. Thus for coding efficiency larger base layers are desired. Larger base layers, however, disqualify more clients from getting the stream. In this paper, we study and quantify the trade-off between the coding efficiency and the range of clients that can be supported. Then, we design an efficient algorithm to compute the optimal size of the base layer that will yield the best video quality for a given client distribution. We implement our algorithm and apply it on video sequences with different characteristics. Our experimental results show that our algorithm improves the average perceived quality for all clients.