Automatic Image Colorization via Multimodal Predictions

Abstract : We aim to color automatically greyscale images, without any manual intervention. The color proposition could then be interactively corrected by user-provided color landmarks if necessary. Automatic colorization is nontrivial since there is usually no one-to-one correspondence between color and local texture. The contribution of our framework is that we deal directly with multimodality and estimate, for each pixel of the image to be colored, the probability distribution of all possible colors, instead of choosing the most probable color at the local level. We also predict the expected variation of color at each pixel, thus defining a nonuniform spatial coherency criterion. We then use graph cuts to maximize the probability of the whole colored image at the global level. We work in the L-a-b color space in order to approximate the human perception of distances between colors, and we use machine learning tools to extract as much information as possible from a dataset of colored examples. The resulting algorithm is fast, designed to be more robust to texture noise, and is above all able to deal with ambiguity, in contrary to previous approaches.
Complete list of metadatas

Cited literature [14 references]  Display  Hide  Download

https://hal.inria.fr/inria-00481552
Contributor : Guillaume Charpiat <>
Submitted on : Friday, May 7, 2010 - 12:19:33 PM
Last modification on : Saturday, August 25, 2018 - 2:48:02 PM
Long-term archiving on : Friday, October 19, 2012 - 2:25:51 PM

File

colorisation.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00481552, version 1

Collections

Citation

Guillaume Charpiat, Matthias Hofmann, Bernhard Schölkopf. Automatic Image Colorization via Multimodal Predictions. European Conference on Computer Vision, Sep 2008, Marseille, France. ⟨inria-00481552⟩

Share

Metrics

Record views

336

Files downloads

738