Deep structured-output regression learning for computational color constancy

The color constancy problem is addressed by structured-output regression on the values of the fully-connected layers of a convolutional neural network. The AlexNet and the VGG are considered and VGG slightly outperformed AlexNet. Best results were obtained with the first fully-connected "fc 6 &...

Full description

Saved in:
Bibliographic Details
Published in2016 23rd International Conference on Pattern Recognition (ICPR) pp. 1899 - 1904
Main Authors Yanlin Qian, Ke Chen, Kamarainen, Joni-Kristian, Nikkanen, Jarno, Matas, Jiri
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.12.2016
Subjects
Benchmark testing
Correlation
Feature extraction
Image color analysis
Lighting
Support vector machines
Training
Online AccessGet full text
DOI10.1109/ICPR.2016.7899914

Cover

More Information
Summary:The color constancy problem is addressed by structured-output regression on the values of the fully-connected layers of a convolutional neural network. The AlexNet and the VGG are considered and VGG slightly outperformed AlexNet. Best results were obtained with the first fully-connected "fc 6 " layer and with multi-output support vector regression. Experiments on the SFU Color Checker and Indoor Dataset benchmarks demonstrate that our method achieves competitive performance, outperforming the state of the art on the SFU indoor benchmark.
DOI:10.1109/ICPR.2016.7899914