Aesthetic Rating and Color Suggestion for Color Palettes

1Japan Advanced Institute of Science and Technology

Pacific Graphics 2016

representative image

Abstract

A model to rate color combinations that considers human aesthetic preferences is proposed. The proposed method does not assume that a color palette has a specific number of colors, i.e., input is not restricted to a two-, three-, or five-color palettes. We extract features from a color palette whose size does not depend on the number of colors in the palette. The proposed rating prediction model is trained using a human color preference dataset. The model allows a user to extend a color palette, e.g., from three colors to five or seven colors, while retaining color harmony. In addition, we present a color search scheme for a given palette and a customized version of the proposed model for a specific color tone. We demonstrate that the proposed model can also be applied to various palette-based applications.

Results

Palette Expansion

Palette Expansion. The original three-color palettes (listed at the center) are expanded with the proposed method. Rows 1-4: a suggested color is added at the index specified by ▼. Row 5: two types of palettes are expanded from three-color to seven-color palettes by adding suggested colors side-by-side.

2D Pattern Coloring

2D Pattern Coloring. The original three-color palette color was expanded to a five-color palette using the colors suggested by the proposed method while retaining its color image, and palette colors are assigned to the pattern templates.

Product Color Design Exploration

Product Color Design Exploration. Left to right: segmented model, model colored by a three-color palette, model colored by a five-color palette, and model colored by a seven-color palette while retaining the color image.

Product Color Variations

Color Variations.

Recoloring

Photo Recoloring. The original photo (left) is recolored using the palette-based photo recoloring method with three- (middle) and seven-color palettes (right).

Downloads

Links

Acknowledgements

We thank all reviewers for their helpful comments. This work was supported by JSPS KAKENHI Grant number 16K12433.

Bibtex

@article{Kita:PG2016,
  title = {Aesthetic Rating and Color Suggestion for Color Palettes},
  author = {Kita, Naoki and Miyata, Kazunori},
  year = {2016},
  journal = {Computer Graphics Forum},
  volume = {35},
  number = {7},
  issn = {1467-8659},
  year = {2016},
  pages = {127--136},
  url = {http://dx.doi.org/10.1111/cgf.13010},
  doi = {10.1111/cgf.13010},
}