Deep multiresolution color constancy

Caglar Aytekin, Jarno Nikkanen, Moncef Gabbouj

    Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

    3 Citations (Scopus)

    Abstract

    In this paper, a computational color constancy method is proposed via estimating the illuminant chromaticity in a scene by pooling from many local estimates. To this end, first, for each image in a dataset, we form an image pyramid consisting of several scales of the original image. Next, local patches of certain size are extracted from each scale in this image pyramid. Then, a convolutional neural network is trained to estimate the illuminant chromaticity per-patch. Finally, two more consecutive trainings are conducted, where the estimation is made per-image via taking the mean (1st training) and median (2nd training) of local estimates. The proposed method is shown to outperform the state-of-the-art in a widely used color constancy dataset.

    Original languageEnglish
    Title of host publication2017 IEEE International Conference on Image Processing, ICIP 2017 - Proceedings
    PublisherIEEE COMPUTER SOCIETY PRESS
    Pages3735-3739
    Number of pages5
    ISBN (Electronic)9781509021758
    DOIs
    Publication statusPublished - 20 Feb 2018
    Publication typeA4 Article in conference proceedings
    EventIEEE International Conference on Image Processing -
    Duration: 1 Jan 1900 → …

    Publication series

    Name
    ISSN (Electronic)2381-8549

    Conference

    ConferenceIEEE International Conference on Image Processing
    Period1/01/00 → …

    Keywords

    • Color constancy
    • Deep learning
    • Illuminant chromaticity estimation
    • Local estimation
    • Multi-resolution

    Publication forum classification

    • Publication forum level 1

    ASJC Scopus subject areas

    • Software
    • Computer Vision and Pattern Recognition
    • Signal Processing

    Fingerprint

    Dive into the research topics of 'Deep multiresolution color constancy'. Together they form a unique fingerprint.

    Cite this