ColorJitter¶
- class mmpretrain.datasets.transforms.ColorJitter(brightness=0.0, contrast=0.0, saturation=0.0, hue=0.0, backend='pillow')[source]¶
Randomly change the brightness, contrast and saturation of an image.
Modified from https://github.com/pytorch/vision/blob/main/torchvision/transforms/transforms.py Licensed under the BSD 3-Clause License.
Required Keys:
img
Modified Keys:
img
- Parameters:
brightness (float | Sequence[float] (min, max)) – How much to jitter brightness. brightness_factor is chosen uniformly from
[max(0, 1 - brightness), 1 + brightness]or the given[min, max]. Should be non negative numbers. Defaults to 0.contrast (float | Sequence[float] (min, max)) – How much to jitter contrast. contrast_factor is chosen uniformly from
[max(0, 1 - contrast), 1 + contrast]or the given[min, max]. Should be non negative numbers. Defaults to 0.saturation (float | Sequence[float] (min, max)) – How much to jitter saturation. saturation_factor is chosen uniformly from
[max(0, 1 - saturation), 1 + saturation]or the given[min, max]. Should be non negative numbers. Defaults to 0.hue (float | Sequence[float] (min, max)) – How much to jitter hue. hue_factor is chosen uniformly from
[-hue, hue](0 <= hue <= 0.5) or the given[min, max](-0.5 <= min <= max <= 0.5). Defaults to 0.backend (str) – The backend to operate the image. Defaults to ‘pillow’