Journal of Signal Processing Systems
Hi-index | 0.00 |
In this paper, the computational complexity of MPEG-4 Fine Granularity Scalability (FGS) coding is analyzed to explore an efficient FGS implementation for video streaming applications. With the proposed hardware-oriented optimization approaches, a hardwired FGS block-level processing core is proposed to provide a cost-effective solution. It can be integrated into an existing MPEG-4 coding system to form an interactive video streaming system. By the proposed FGS coder with reordered coding flow, the streaming server can adaptively decide quality-enhanced region by selective enhancement according to both object information from encoding side and user-defined region from receiver side. The proposed hardware core can support FGS profile level 5, frame size 720x576, 30Hz, for real-time streaming applications at 54 MHz.