Abstract: Multi-view video is a kind of video sequence taken by a set of cameras. Generally, it is difficult for the performance parameter of each camera to totally agree with each other, which consequently causes color differences among video images of different viewpoints. A new color compensation model is proposed for the Multi-View Video Coding（MVC）, which can efficiently use information of inter-view of mutual independent coding to reduce the color mismatch caused by multi-arrays of camera shooting. This method can establish color compensation model in RGB color space and furthermore can extend it to YCbCr color space to suit for present video code standard. Compared to the previous inter-view reference images, the modified inter-view reference pictures are more similar to the present code image which consequently improves the coding efficiency of MVC. Moreover, the color compensation parameter can be derived in both encoder and decoder, therefore no additional data need to be transmitted to the decoder. The experimental results also show that the proposed method improves the coding efficiency of MVC and obtains a better video quality.

Key words: Multi-View Video Coding（MVC）, H.264/AVC, color compensation, diversity of cameras

摘要： 多视点视频是由一组摄像机拍摄得到的视频序列。通常情况下，各个摄像机的性能参数很难做到完全一致，从而引起不同视点的视频图像之间产生颜色差异。提出了在多视点视频编码中使用一种新的颜色补偿模式，这种模式能够有效利用相互独立编码的帧间视频的信息，从而减少由多阵列摄像而带来的颜色误匹配。该方法在RGB颜色空间上建立颜色补偿模型，并进一步将颜色补偿模型扩展到YCbCr颜色空间以适用现有视频编码标准。与原始视点间参考图像相比，经过补偿后的视点间参考图像与当前编码图像更为相似，从而提高了MVC的编码效率。此外，颜色补偿模型参数可以分别从编码端和解码端推导得出，不需要将其写入码流。实验结果也表明提出的方法能够提高编码的效率和得到更好的视频质量。

关键词: 多视点视频编码（MVC）, H.264/AVC, 颜色补偿, 分布式摄像