An enhanced fast mode decision model for spatial enhancement layers in scalable video coding

Generic techniques to reduce SVC enhancement layer encoding complexity

“…Based on these observations, the authors have proposed a method which also takes the quantization of both layers into account [11], achieving a complexity reduction of 75%, while maintaining a high RD performance. Modes with a high probability for being selected in the enhancement layer are always evaluated.…”

“…Some of these ideas have been used in [11]. However, general sub-mode decision optimizations will significantly improve any existing fast mode decision models.…”

Computationally efficient and adaptive scalable video coding

“…Recently, several fast mode decision algorithms [114][115][116][117][118][119] were proposed to reduce encoder complexity with a focus on scalable video coding. Many of them cited our work and presented improved performance as follows.…”

“…• Similar to our proposed algorithm, Sebastiaan et al explored the correlation between the macroblock mode in the enhancement layer and the co-located macroblock in the reference layer in [118]. This observation was used in conjunction with the quantization of the enhancement layer to design a fast mode decision model for dyadic and non-dyadic spatial enhancement layers.…”

“…As such, we have used Lin's method as a benchmark for CGS scalability in order to have a common basis for performance comparison. Nevertheless, we also paid attention to recently-published methods, for example, Sebastiaan's method [118], which attracted much attention due to its effective complexity reduction and easy implementation. Therefore, we compared In order to compare the performance of our proposed algorithm versus Lin's method, we implemented both algorithms in the JSVM encoder.…”