I am implementing the application of a ResNet neural network across different color spaces.
I apply masks to the training images that delineate the area to be left for the network to learn.
The question is whether applying conversion to a different color space for an already normalized RGB image is ok? For example, normalized RGB image convert to HSV and then pass it to CNN.
Should I first subject the image to conversion and then normalization?
I am using that method for calculate mean and std for my dataset: https://www.binarystudy.com/2021/04/how-to-calculate-mean-standard-deviation-images-pytorch.html
Can I use this method to calculate values for normalization in other color spaces?
The other color spaces I'm using are HSV, HSL, YUV, LAB, YCrCb, etc.
I am implementing the application of a ResNet neural network across different color spaces.
I apply masks to the training images that delineate the area to be left for the network to learn.
The question is whether applying conversion to a different color space for an already normalized RGB image is ok? For example, normalized RGB image convert to HSV and then pass it to CNN.
Should I first subject the image to conversion and then normalization?
I am using that method for calculate mean and std for my dataset: https://www.binarystudy.com/2021/04/how-to-calculate-mean-standard-deviation-images-pytorch.html
Can I use this method to calculate values for normalization in other color spaces?
The other color spaces I'm using are HSV, HSL, YUV, LAB, YCrCb, etc.