Universal Image Quality Index¶

Module Interface¶

class torchmetrics.UniversalImageQualityIndex(kernel_size=(11, 11), sigma=(1.5, 1.5), reduction='elementwise_mean', data_range=None, **kwargs)[source]

Computes Universal Image Quality Index (UniversalImageQualityIndex).

As input to forward and update the metric accepts the following input

preds (Tensor): Predictions from model of shape (N,C,H,W)
target (Tensor): Ground truth values of shape (N,C,H,W)

As output of forward and compute the metric returns the following output

uiqi (Tensor): if reduction!='none' returns float scalar tensor with average UIQI value over sample else returns tensor of shape (N,) with UIQI values per sample

Parameters

kernel_size¶ (Sequence[int]) – size of the gaussian kernel
sigma¶ (Sequence[float]) – Standard deviation of the gaussian kernel
reduction¶ (Literal[‘elementwise_mean’, ‘sum’, ‘none’, None]) –
a method to reduce metric score over labels.
- 'elementwise_mean': takes the mean (default)
- 'sum': takes the sum
- 'none' or None: no reduction will be applied
data_range¶ (Optional[float]) – Range of the image. If None, it is determined from the image (max - min)
kwargs¶ (Any) – Additional keyword arguments, see Advanced metric settings for more info.

Returns

Tensor with UniversalImageQualityIndex score

Example

>>> import torch
>>> from torchmetrics import UniversalImageQualityIndex
>>> preds = torch.rand([16, 1, 16, 16])
>>> target = preds * 0.75
>>> uqi = UniversalImageQualityIndex()
>>> uqi(preds, target)
tensor(0.9216)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

Functional Interface¶

torchmetrics.functional.universal_image_quality_index(preds, target, kernel_size=(11, 11), sigma=(1.5, 1.5), reduction='elementwise_mean', data_range=None)[source]

Universal Image Quality Index.

Parameters

preds¶ (Tensor) – estimated image
target¶ (Tensor) – ground truth image
kernel_size¶ (Sequence[int]) – size of the gaussian kernel
sigma¶ (Sequence[float]) – Standard deviation of the gaussian kernel
reduction¶ (Optional[Literal[‘elementwise_mean’, ‘sum’, ‘none’]]) –
a method to reduce metric score over labels.
- 'elementwise_mean': takes the mean (default)
- 'sum': takes the sum
- 'none' or None: no reduction will be applied
data_range¶ (Optional[float]) – Range of the image. If None, it is determined from the image (max - min)

Return type

Tensor

Returns

Tensor with UniversalImageQualityIndex score

Raises

TypeError – If preds and target don’t have the same data type.
ValueError – If preds and target don’t have BxCxHxW shape.
ValueError – If the length of kernel_size or sigma is not 2.
ValueError – If one of the elements of kernel_size is not an odd positive number.
ValueError – If one of the elements of sigma is not a positive number.

Example

>>> from torchmetrics.functional import universal_image_quality_index
>>> preds = torch.rand([16, 1, 16, 16])
>>> target = preds * 0.75
>>> universal_image_quality_index(preds, target)
tensor(0.9216)

References

[1] Zhou Wang and A. C. Bovik, “A universal image quality index,” in IEEE Signal Processing Letters, vol. 9, no. 3, pp. 81-84, March 2002, doi: 10.1109/97.995823.

[2] Zhou Wang, A. C. Bovik, H. R. Sheikh and E. P. Simoncelli, “Image quality assessment: from error visibility to structural similarity,” in IEEE Transactions on Image Processing, vol. 13, no. 4, pp. 600-612, April 2004, doi: 10.1109/TIP.2003.819861.