Narrow Results

Color transfer is to alter an image’s color composition by reference to the color characteristics of another image. In this paper, we build a system called artistic coloring that realizes automatic color transfer from famous paintings. It properly extracts the wonderful color characteristics of famous paintings and applies them to color transfer. Specially, we investigate the traditional color theme extraction methods and find their deficiencies. Based on this, we quantify the processing of human painting to find the main colors in the color palette and propose an artistic balanced color theme extraction algorithm aimed specially at paintings. In the experiments, a user study is carried out to evaluate the artistic balanced extraction method. Our proposed method achieves the highest score 3.98 in a 5-point evaluation, which is 41% higher than traditional methods at most. We have successfully tested the artistic coloring system on lots of images with different painting styles. The results are natural and have the similar color characteristics with their corresponding reference paintings.

This paper proposes a flexible example-based color transfer system, providing an automatic mode and an advanced mode, for novices and expert users, respectively. The experimental results show that the proposed color transfer system not only can introduce natural results with simple operation by novices in the first use, but also can produce various desired results by users with short-term learning.

The paper proposes a novel algorithm based on Retinex theory and color transfer to get illumination invariance among images, taking one of the images as a reference. Most of the algorithms focus on eliminating the illumination effect of one image, while our method corrects the variable illumination among images. Therefore, the algorithm can be applied to target tracking, recognition, matching and 3D reconstruction to minish the illumination variations. Experiment results validate the method.

Color transfer means desired look of source image is determined by reference color image. Obviously, a correct mapping between source image and reference image is key problem for color transfer. However, natural color image contains a lot of different color, texture and other features, which brings difficulty for color transfer. In this paper, we propose a new color transfer method based on patch similarity metric. All patches used for similarity mapping is centered on feature points extracted with a Gaussian pyramid iterative approach. The mapping between source patch and reference patch is resolved by spatial pyramid matching. Finally, the color of feature points is updated by the mapped points in reference image. Then, a global patch driven colorization is applied for source image to obtain final color transfer result. Experimental results highlight our superior performance of proposed color transfer method.