Color transfer in correlated color space

In this paper we present a process called color transfer which can borrow one image's color characteristics from another. Recently Reinhard and his colleagues reported a pioneering work of color transfer. Their technology can produce very believable results, but has to transform pixel values from RGB to lαβ. Inspired by their work, we advise an approach which can directly deal with the color transfer in any 3D space.From the view of statistics, we consider pixel's value as a three-dimension stochastic variable and an image as a set of samples, so the correlations between three components can be measured by covariance. Our method imports covariance between three components of pixel values while calculate the mean along each of the three axes. Then we decompose the covariance matrix using SVD algorithm and get a rotation matrix. Finally we can scale, rotate and shift pixel data of target image to fit data points' cluster of source image in the current color space and get resultant image which takes on source image's look and feel. Besides the global processing, a swatch-based method is introduced in order to manipulate images' color more elaborately. Experimental results confirm the validity and usefulness of our method.

[1]  Dani Lischinski,et al.  Colorization by example , 2005, EGSR '05.

[2]  Harry Shum,et al.  Bayesian Correction of Image Intensity with Spatial Consideration , 2004, ECCV.

[3]  Yoav Y. Schechner,et al.  Clear underwater vision , 2004, CVPR 2004.

[4]  Dani Lischinski,et al.  Colorization using optimization , 2004, ACM Trans. Graph..

[5]  Konstantinos Konstantinides,et al.  Statistical analysis of effective singular values in matrix rank determination , 1988, IEEE Trans. Acoust. Speech Signal Process..

[6]  James A. Ferwerda,et al.  Applied perception , 2001, IEEE Computer Graphics and Applications.

[7]  Erik Reinhard,et al.  Color Transfer between Images , 2001, IEEE Computer Graphics and Applications.

[8]  Lily Lee,et al.  Gait analysis for classification , 2002 .

[9]  Francesca Gasparini,et al.  Color correction for digital photographs , 2003, 12th International Conference on Image Analysis and Processing, 2003.Proceedings..

[10]  Yoav Y. Schechner,et al.  Clear underwater vision , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[11]  Klaus Mueller,et al.  Transferring color to greyscale images , 2002, ACM Trans. Graph..

[12]  D. Ruderman,et al.  Statistics of cone responses to natural images: implications for visual coding , 1998 .

[13]  Bruce Gooch,et al.  Color2Gray: salience-preserving color removal , 2005, ACM Trans. Graph..

[14]  Dani Lischinski,et al.  Colorization using optimization , 2004, SIGGRAPH 2004.

[15]  T. Watanabe,et al.  High quality color correction method combining neural networks with genetic algorithms , 2001, Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205).