Abstract
We develop and evaluate motion compensation schemes for predictive coding of the component color television signal. Algorithms are discussed for estimation of motion of each color component (luminance and chrominance) separately as well as in combination. Techniques for switching of the predictors for individual components are proposed and simulated. Simulations show that it is sufficient to estimate parameters of motion based only on the luminance and use them for motion-based prediction and switching the predictors for both the luminance and chrominance. Thus, only one motion estimator and prediction switch is needed for the three components of the color signal. Compression capability of motion compensation is scene dependent, in some video conference type of scenes, bit rate is reduced by as much as 60 percent compared to conditional replenishment coding.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
