Inference on JPG and PNG images

[tejgvsl]

Hi Mr. Hu,
When I try to run the inference on illuminated .PNG or .jpg images, observed the illumination correction is not as good as for the images from the mentioned datasets. Observing a pale pink tint in the corrected output image for various samples.

So does it mean the trained model works only for linear images and in order to make it work for non linear images (jpg, Png) does it need to be re-trained?

[mpaden3]

The way color constancy CNN models (including this one) work is that they learn how to correct color that camera sensor produces (camera sensitivity function), as well as color from the source of illumination. Therefore, a particular model is only usable on the type of images it was trained on, in this case unprocessed images taken by a particular camera sensor.
Universal estimator using CNN, would require a prestep to remove the impact of camera sensor. Interesting idea can be found in this paper.
https://arxiv.org/pdf/1912.06888.pdf