Digital watermarking for real-time hardware accelerated scalable video streaming

To sustainably address these high computational complexities, this research proposes a low complexity and a highly transparent watermarking technique for real-time hardware accelerated spatial scalable video streaming. In achieving this goal, two contributions are discussed here. The first contributi...

全面介紹

Saved in:
書目詳細資料
主要作者: Buhari, Adamu Muhammad
格式: Thesis
出版: 2015
主題:
標簽: 添加標簽
沒有標簽, 成為第一個標記此記錄!
實物特徵
總結:To sustainably address these high computational complexities, this research proposes a low complexity and a highly transparent watermarking technique for real-time hardware accelerated spatial scalable video streaming. In achieving this goal, two contributions are discussed here. The first contribution puts forward a real-time high resolution down-sampling algorithm on many-core processor for spatial scalable video coding. This contribution analyses the principal architecture of a serial down-sampling algorithm in the Joint-Scalable-Video-Model (JSVM) reference software and identify its performance limitations for spatial scalability. Then, a parallel multi-core down-sampling algorithm is studied as benchmark. Experimental results for this algorithm using an 8-core processor exhibits performance speedup of 5.25× in down-sampling a QXGA video resolution into three lower resolution layers (i.e., Full-HD, HD and quarter-HD).