Dominant Color Tracking Based Color Correction for Multi-View Video Using Kalman Filter

Multi-view video is a new multimedia service which provides interactive experience and depth perception. Color variations between the camera views are necessary to be eliminated in multi-view imaging. In this paper, a dominant color tracking based color correction method for multi-view video is proposed using Kalman filter. It tracks the dominant colors by using Kalman filter to fit the color variation of the real corrected multi-view videos. Experimental results show that the proposed method can eliminate discontinuous variations between frames for higher coding performance.