Deep color transfer using histogram analogy
- PDF / 6,165,679 Bytes
- 15 Pages / 595.276 x 790.866 pts Page_size
- 102 Downloads / 298 Views
ORIGINAL ARTICLE
Deep color transfer using histogram analogy Junyong Lee1 · Hyeongseok Son1 · Gunhee Lee2 · Jonghyeop Lee1 · Sunghyun Cho1 · Seungyong Lee1
© Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract We propose a novel approach to transferring the color of a reference image to a given source image. Although there can be diverse pairs of source and reference images in terms of content and composition similarity, previous methods are not capable of covering the whole diversity. To resolve this limitation, we propose a deep neural network that leverages color histogram analogy for color transfer. A histogram contains essential color information of an image, and our network utilizes the analogy between the source and reference histograms to modulate the color of the source image with abstract color features of the reference image. In our approach, histogram analogy is exploited basically among the whole images, but it can also be applied to semantically corresponding regions in the case that the source and reference images have similar contents with different compositions. Experimental results show that our approach effectively transfers the reference colors to the source images in a variety of settings. We also demonstrate a few applications of our approach, such as palette-based recolorization, color enhancement, and color editing. Keywords Color transfer · Histogram analogy · Photo-realistic style transfer · Recolorization
1 Introduction Color transfer is the task of converting a source image according to the reference color information that has desired color characteristics. Many previous methods use guidance images for the reference color information, where the reference images with similar contents and compositions to the source images are needed to obtain visually pleasing results. Such a setting would not be ideal for users as an additional task of searching desired images is required. In this paper, we focus
B
Seungyong Lee [email protected] Junyong Lee [email protected] Hyeongseok Son [email protected] Gunhee Lee [email protected] Jonghyeop Lee [email protected] Sunghyun Cho [email protected]
1
Department of Computer Science and Engineering, POSTECH, Pohang, South Korea
2
NCSOFT, Seongnam, South Korea
on handling various reference images so that our method can produce plausible color transfer results for diverse pairs of source and reference images (Fig. 1). We first divide various correlations between source and reference images into three cases (Fig. 1). The first case is the strong relevance, where two images have high similarity in both the contents and positions of semantic objects. Second, the weak relevance refers to high similarity in the contents but with less correlations in the object spatial configurations. The last case of irrelevance includes image pairs with dissimilar contents and special settings with graphic design images and color palettes as the reference. Traditional global color transfer methods (GCT) [5,29, 30], which transfer th
Data Loading...
