SCT: Superpixel-based Color Transfer


Back to research page

Highlights


  • New Superpixel-based Color Transfer (SCT) method between a target and a source image
  • New constraint to enforce the match diversity of a fast superpixel matching [2]
  • Color fusion of the selected superpixels, with respect to the initial grain and exposure of the target image
  • Results obtained in less than 1s, and more satisfying than the ones of state-of-the-art methods

Method


Superpixel decomposition
  • Fast and accurate superpixel decomposition using SCALP [2]




Constrained superpixel matching
  • Fast superpixel matching between the target and the source image using [3]
  • To enforce the match diversity: Limitation of the selection of source superpixels to ε times:


Without constraint (ε=∞), the palette of the source image is not catched
With the constraint (ε=1), the target superpixels are forced to catch other than red colors


Color fusion
  • Color fusion framework inspired from non-local means [4] and based on color and spatial similarity





Impact of the ε constraint on match diversity
  • The ε constraint enables to globally catch the palette of the source image

The maps (bottom row) indicate the selection number of source superpixels (black is zero, white is the highest selection number)


Comparison to state-of-the-art methods
  • Results equivalent or more satisfying than the ones of state-of-the-art methods





Main (to cite)

  1. R. Giraud, V.-T. Ta and N. Papadakis: Superpixel-based Color Transfer
    IEEE International Conference on Image Processing (ICIP), 2017   https://hal.archives-ouvertes.fr/hal-01519644/file/Giraud_SCT_ICIP17.pdf   BibTex    
Others

  1. R. Giraud, V.-T. Ta and N. Papadakis: Transfert de couleurs basé superpixels
    Groupe d'Études du Traitement du Signal et des Images (GRETSI), 2017   https://hal.archives-ouvertes.fr/hal-01542596/file/Giraud_SCT_GRETSI_2017.pdf   BibTex    


  1. R. Giraud, V.-T. Ta and N. Papadakis: SCALP: Superpixels with Contour Adherence using Linear Path
    International Conference on Pattern Recognition (ICPR), 2016
  2. R. Giraud, V.-T. Ta, A. Bugeau, et al.: SuperPatchMatch: an Algorithm for Robust Correspondences using Superpixel Patches
    IEEE Trans. on Image Processing (TIP), 2017
  3. A. Buades, B. Coll, and J.-M. Morel: A non-local algorithm for image denoising
    IEEE Int. Conf. on Computer Vision and Pattern Recognition (CVPR), 2005
  4. F. Pitié, A. Kokaram, and R. Dahyot: Automated colour grading using colour distribution transfer
    Computer Vision and Image Understanding (CVIU), 2007
  5. N. Papadakis, E. Provenzi, and V. Caselles: A variational model for histogram transfer of color images
    IEEE Trans. on Image Processing (TIP), 2011
  6. R. Nguyen, S. J. Kim, and M. S. Brown: Illuminant aware gamut-based color transfer
    Comput. Graph. Forum, 2014


Back to top