Study of Very Low Complexity Image Coding via Compressive Sampling and Colorization

点击翻译按钮获取中文摘要

We designed two research tracks for developing new very low complexity image coding by using colorization method and compressive sampling. 1) New colorization based coding scheme without sending the positions of color information is studied. The positions can be estimated by grayscale image at the decoder side and local decoding allows us to obtain which position will be estimated. 2 It is well known that the texture component of natural image is difficult to compress. Our research reveals that, by utilizing the prior information about texture component, we can decode the texture component precisely from the noiselet coefficient. From these results, we successfully developed the component technologies for establishing the very low complexity image coding.