YUV Correction for Multi-View Video Compression

A luminance and color correction algorithm for multi-view video compression is proposed in this paper. Due to the dissimilar radiometric characteristics of different cameras and the variation of lighting conditions, significant luminance and color discrepancies among different camera views often exist in multi-view video sequences. These variations may cause the wrong disparity matching and result in the degradation of coding efficiency. To elevate the multi-view video coding efficiency, a discrepancy model of different views based on modulation and translation parameters is proposed, and the discrepancies among different views are compensated before disparity-matching. The optimization of compensation parameters is discussed, and histogram matching method is used to mitigate the computational load. Experimental results on both visual quality and coding efficiency have verified the validity and robustness of the proposed method