In a stereo camera set-up for 3D cinema, much care is taken to ensure/nthat the cameras are both the same model and use identical/nparameter values. But an imperfect adjustment, the use of beam/nsplitters, as well as unavoidable differences in the characteristics/nof the cameras and lens systems, often cause visible differences in/ncolor among the two views. Many of these differences appear to/nbe local and therefore cannot be fully removed by common color/nmatching approaches, which are typically ...
In a stereo camera set-up for 3D cinema, much care is taken to ensure/nthat the cameras are both the same model and use identical/nparameter values. But an imperfect adjustment, the use of beam/nsplitters, as well as unavoidable differences in the characteristics/nof the cameras and lens systems, often cause visible differences in/ncolor among the two views. Many of these differences appear to/nbe local and therefore cannot be fully removed by common color/nmatching approaches, which are typically global methods. We propose/na simple yet effective method for local color matching, that/noperates in three steps. First, one view is chosen as the target and/nmorphed so that it becomes locally registered with the source view,/nthe one whose colors will be modified. Next, color matching is performed/non the source view by aligning each local histogram with/nthe corresponding local histogram in the warped target view. Finally,/nPoisson editing [9] is applied on the source regions which/nhave no correspondence in the target. For several professional, high/nquality examples of 3D sequences, we show that our algorithm outperforms/nfour different conventional color matching approaches.
+
