Model-based color manipulation

  • Authors:
  • Lior Shapira;Daniel Cohen-Or;Lior Gavish;Ariel Shamir

  • Affiliations:
  • Tel-Aviv University;Tel-Aviv University;Tel-Aviv University;Interdisciplinary Center, Herzliya

  • Venue:
  • ACM SIGGRAPH 2007 posters
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

The manipulation of the colors in an image and color spaces in general can be highly complex and non-intuitive. Converting the thoughts or vision of the designer to specific color modifications on an image is hard, while envisioning the effect of a specific color manipulation is difficult as well. Furthermore, recoloring must take into account spatial image considerations along with the color constraints. We introduce a method that models possible color manipulations of an image by a structured search space. We utilized the design galleries metaphor to present the designer with an interface for navigating this search space toward a desired target color variation. The pixels' colors are modeled by a spatially-constrained Gaussian mixture model of the chroma channels, and the search space is defined by applying affine transformations to the parameters of this model. This approach provides both inspiration and intuitive navigation in the complex space of image color manipulations.