Moving-picture layered coding and decoding methods, apparatuses, and programs

摘要

An input video signal is encoded at a plurality of coding layers exhibiting different spatial resolutions. Decoded is a given signal coded at a lower coding layer lower than a specific coding layer among the plurality of coding layers to generate a decoded signal of the lower coding layer. Spatial interpolation is applied to the decoded signal of the lower coding layer to generate an upscaled decoded video signal. The spatial interpolation is an upscaling procedure to upscale the decoded signal of the lower coding layer into a spatial resolution of the specific coding layer. A spatial high-frequency components estimation and scale up procedure is applied to the decoded signal of the lower coding layer to generate a high-frequency components signal. The upscaled decoded video signal and the high-frequency components signal are subtracted from the input video signal exhibiting a spatial resolution of the specific coding layer to produce a predictive-error signal. Then, the predictive-error signal is encoded at the specific coding layer to generate an output bitstream. The decoding, spatial interpolating, spatial high-frequency components estimating and upscaling, and encoding procedures are executed at, at least, the specific coding layer among the plurality of coding layers except the lowest coding layer exhibiting the lowest spatial resolution thereamong.