Learning Representations for Automatic Colorization

Abstract

We develop a fully automatic image colorization system. Our approach leverages recent advances in deep networks, exploiting both low-level and semantic representations. As many scene elements naturally appear according to multimodal color distributions, we train our model to predict per-pixel color histograms. This intermediate output can be used to automatically generate a color image, or further manipulated prior to image formation. On both fully and partially automatic colorization tasks, we outperform existing methods. We also explore colorization as a vehicle for self-supervised visual representation learning.

Cite

Text

Larsson et al. "Learning Representations for Automatic Colorization." European Conference on Computer Vision, 2016. doi:10.1007/978-3-319-46493-0_35

Markdown

[Larsson et al. "Learning Representations for Automatic Colorization." European Conference on Computer Vision, 2016.](https://mlanthology.org/eccv/2016/larsson2016eccv-learning/) doi:10.1007/978-3-319-46493-0_35

BibTeX

@inproceedings{larsson2016eccv-learning,
  title     = {{Learning Representations for Automatic Colorization}},
  author    = {Larsson, Gustav and Maire, Michael and Shakhnarovich, Gregory},
  booktitle = {European Conference on Computer Vision},
  year      = {2016},
  pages     = {577-593},
  doi       = {10.1007/978-3-319-46493-0_35},
  url       = {https://mlanthology.org/eccv/2016/larsson2016eccv-learning/}
}