Global contrast normalization in Pytorch

Hi all, how can I implement global contrast normalization in the form of ‘transforms.Compose’? any help would be appreciated.

You probably want just the standard ImageNet style RMS normalization? I.e. transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225)) or similar…

1 Like

I want to apply the global contrast normalization to the CIFAR-10 dataset. This type of normalization was investigated by “Goodfellow, Ian J, Warde-Farley, David, Mirza, Mehdi, Courville, Aaron, and Bengio, Yoshua. Maxout networks. arXiv preprint arXiv:1302.4389, 2013.”