Avoid common mistakes on your manuscript.
1 Erratum to: Vis Comput (2014) 30:685–695 DOI 10.1007/s00371-014-0962-1
The original version of this article unfortunately contained two mistakes.
In the second paragraph of Sect. 4.2 “Refinement with skin color model”, the second sentence after Eq. (9) should read:
Here \(\mathbf{R}_\mathrm{k}\) denoting the reliable skin tone is just computed as follows: Sort the pixels in non-noise areas of one side (left or right side) by their skin color likelihoods in descending order then we can get \(\mathbf{R}_\mathrm{k}\) by striking the average reflectance value of the first K pixels.
The end of the last paragraph of Sect. 5 “Experiment”, it should read:
As a comparison, we also show the result of directly transferring the shading of the reference images to the input images without decomposition [10]. It is obvious that our results are much more realistic and natural.
Author information
Authors and Affiliations
Corresponding author
Additional information
The online version of the original article can be found under doi:10.1007/s00371-014-0962-1.
Rights and permissions
About this article
Cite this article
Hu, K., Liu, Y., Dong, Q. et al. Erratum to: Color face image decomposition under complex lighting conditions. Vis Comput 30, 1069 (2014). https://doi.org/10.1007/s00371-014-1004-8
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00371-014-1004-8