Effective cross-sensor color constancy using a dual-mapping strategy

Shuwei Yue and *Minchen Wei

Color, Imaging, and Metaverse Research Center, The Hong Kong Polytechnic University.

BibTex

Please cite us if you use this code or our paper:

@article{Yue24,
author = {Shuwei Yue and Minchen Wei},
journal = {J. Opt. Soc. Am. A}, number = {2},
pages = {329--337},
publisher = {Optica Publishing Group},
title = {Effective cross-sensor color constancy using a dual-mapping strategy},
volume = {41},
month = {Feb},
year = {2024}
}

PDF

PDF(Manuscript )

Code

Prerequisite

• Pytorch
• opencv-python

Training

Choose the training sensor and testing sensor in get_args()function of train.py

run python train.py

Testing

Define the testing sensor and selecting the related pre-trained model

run python test.py

Frequent questions

1. How to train your own dataset

• step1: generate the diagonal matrix(need a white point under 6500K as described in our paper)
• step2: prepare your own testing data

2. Sample code

   TBD