Normalize

class mmpretrain.datasets.transforms.Normalize(mean, std, to_rgb=True)

Normalize the image.

Required Keys:

• img

Modified Keys:

• img

Added Keys:

• img_norm_cfg

  • mean

  • std

  • to_rgb

Parameters:

• mean (sequence) – Mean values of 3 channels.

• std (sequence) – Std values of 3 channels.

• to_rgb (bool) – Whether to convert the image from BGR to RGB before normlizing the image. If to_rgb=True, the order of mean and std should be RGB. If to_rgb=False, the order of mean and std should be the same order of the image. Defaults to True.

transform(results)

Function to normalize images.

Parameters:

results (dict) – Result dict from loading pipeline.

Returns:

Normalized results, key ‘img_norm_cfg’ key is added in to result dict.

Return type:

dict