HDR CFA Image Rendering


David Alleysson, Laurence Meylan and Sabine Süsstrunk


We propose a method for high dynamic range (HDR) mapping that is directly applied on the color filter array (CFA) image instead of the already demosaiced image. This rendering is closer to retinal processing where an image is acquired by a mosaic of cones and where adaptive non-linear functions apply before interpolation. Thus, in our framework, demosaicing is the final step of the rendering. Our method, inspired by retinal sampling and adaptive processing is very simple, fast because only one third of operations are needed, and gives good result as shown by experiments.

Matlab code is available upon request.


Figure 4: Example of image processing (Top left) The image is solely demosaiced (Top right) The first non linearity and the demosaicing process (Bottom) two non-linearities followed by demosaicing.

CRW_0403demos.jpg CRW_0403horizdemos.jpg

Figure 5: Comparison between several methods (Top left) method of [15] (Top right) method of [7] (Bottom) our presented method.

memorialmeylan.jpg memorialdurand.jpg

Figure 6: Comparison between several methods (Top left) method in [15] (Top right) our method (Bottom left) method in [7] (Bottom right) our method.

Laurencecar.jpg CFAcar.jpg
DurandRoberto.jpg CFARoberto.jpg


[7] F. Durand and J. Dorsey. “Fast bilateral filtering for the display of high-dynamic-range images”. Proc, SIGGRAPH 02. pp. 257-266, 2002.
[15] L. Meylan and S. Susstrunk, “High Dynamic Range Image Rendering Using a Retinex-Based Adaptive Filter”,to appear in IEEE Transactions on Image Processing, 2006.

Supplementary Figures

Left: gamma-encoded image. Right: image rendered with our method.

bristolb.jpg bristolbCFA.jpg
On.jpg OnCFA.jpg
DSC59.jpg DSC59_CFA.jpg