Combined Inter-frame and Inter-color Prediction for Color Video Denoising

An advanced color video denoising scheme which we call CIFIC based on combined inter-frame and inter-color prediction is presented in this paper. CIFIC performs the denoising filtering in the RGB color space, and exploits both the inter-frame and inter-color correlation in color video signal directly by forming multiple predictors for each color component using all three color components in the current frame as well as the motion-compensated neighboring reference frames. The temporal correspondence is established through the joint-RGB noise-robust motion estimation (ME) which acquires a single motion trajectory for the RGB components. Then the current noisy observation as well as the inter-frame and inter-color predictors are combined by a linear minimum mean squared error (LMMSE) filter to obtain the denoised estimate for every color component. The experimental results verify that CIFIC provides competitive performance both in terms of the objective metric and in perceptual quality when compared with other state-of-the-art algorithms.