BIG

Bristol Interaction and Graphics

Histogram Reshaping

Abstract


Image manipulation takes many forms. A powerful approach involves image adjustment by example. To make color edits more intuitive, the intelligent transfer of a user-specified target image's color palette can achieve a multitude of creative effects, provided the user is supplied with a small set of straightforward parameters. We present a novel histogram reshaping technique which allows significantly more control than previous methods. Given that the user is free to chose any image as the target, the process of steering the algorithm becomes artistic. Moreover, we show for the first time that creative tone reproduction can be achieved by matching a high dynamic range image against a low dynamic range target.

Details


Paper: Tania Pouli, Erik Reinhard, 'Progressive Color Transfer for Images of Arbitrary Dynamic Range', Computers and Graphics 35(1), pp. 67-80, Elsevier, 2011.
Paper: Tania Pouli, Erik Reinhard, 'Progressive Histogram Reshaping for Creative Color Transfer and Tone Reproduction', International Symposium on Non-Photorealistic Animation and Rendering (NPAR), published by ACM, 2010, pp. 81-90. 

Results

HDR to LDR color transfer results

HDR to LDR color transfer results. A variety of images were chosen as the target, achieving very different styles.

Partial match

Here is an example of a partial result. Our algorithm allows the user to control how much of the style of the target image should be adopted.

Color transfer using masks

To further control the result of the transfer, a mask can be specified for both the source and the target to define which regions should be considered (either when computing the colors or when deciding which pixels should be changed).