Recurrent Color Constancy

Yanlin Qian, Ke Chen, Jarno Nikkanen, Joni-Kristian Kamarainen, Jiri Matas

ICCV 2017

doi:10.1109/ICCV.2017.582 /iccv/2017/qian2017iccv-recurrent/

Abstract

We introduce a novel formulation of temporal color constancy which considers multiple frames preceding the frame for which illumination is estimated. We propose an end-to-end trainable recurrent color constancy network -- the RCC-Net -- which exploits convolutional LSTMs and a simulated sequence to learn compositional representations in space and time. We use a standard single frame color constancy benchmark, the SFU Gray Ball Dataset, which can be adapted to a temporal setting. Extensive experiments show that the proposed method consistently outperforms single-frame state-of-the-art methods and their temporal variants.

PDF ICCV Semantic Scholar

Cite

Text

Qian et al. "Recurrent Color Constancy." International Conference on Computer Vision, 2017. doi:10.1109/ICCV.2017.582

Markdown

[Qian et al. "Recurrent Color Constancy." International Conference on Computer Vision, 2017.](https://mlanthology.org/iccv/2017/qian2017iccv-recurrent/) doi:10.1109/ICCV.2017.582

BibTeX

@inproceedings{qian2017iccv-recurrent,
  title     = {{Recurrent Color Constancy}},
  author    = {Qian, Yanlin and Chen, Ke and Nikkanen, Jarno and Kamarainen, Joni-Kristian and Matas, Jiri},
  booktitle = {International Conference on Computer Vision},
  year      = {2017},
  doi       = {10.1109/ICCV.2017.582},
  url       = {https://mlanthology.org/iccv/2017/qian2017iccv-recurrent/}
}