pyiqa.archs.msswd_arch
======================

.. py:module:: pyiqa.archs.msswd_arch

.. autoapi-nested-parse::

   Perceptual color difference metric, MS-SWD.

   @inproceedings{he2024ms-swd,
     title={Multiscale Sliced {Wasserstein} Distances as Perceptual Color Difference Measures},
     author={He, Jiaqi and Wang, Zhihua and Wang, Leon and Liu, Tsein-I and Fang, Yuming and Sun, Qilin and Ma, Kede},
     booktitle={European Conference on Computer Vision},
     pages={1--18},
     year={2024},
     url={http://arxiv.org/abs/2407.10181}
   }

   Reference:
       - Official github: https://github.com/real-hjq/MS-SWD







Module Contents
---------------

.. py:function:: color_space_transform(input_color, fromSpace2toSpace)

   Transform color tensors between supported color spaces.

   :param input_color: Color tensor with shape ``(N, C, H, W)``.
   :type input_color: torch.Tensor
   :param fromSpace2toSpace: Conversion key, for example ``'srgb2lab'``.
   :type fromSpace2toSpace: str

   :returns: Transformed tensor with shape ``(N, C, H, W)``.
   :rtype: torch.Tensor

   :raises ValueError: If the conversion key is not defined.


.. py:class:: MS_SWD_learned(resize_input: bool = True, pretrained: bool = True, pretrained_model_path: str = None, **kwargs)

   Bases: :py:obj:`torch.nn.Module`


   MS-SWD perceptual color difference metric.

   :param resize_input: Whether to resize inputs with short side larger
                        than ``256`` before scoring.
   :type resize_input: bool
   :param pretrained: Whether to load pretrained weights.
   :type pretrained: bool
   :param pretrained_model_path: Optional local checkpoint path.
   :type pretrained_model_path: str | None
   :param \*\*kwargs: Reserved compatibility arguments.


   .. py:method:: preprocess_img(x)

      Optionally resize image batch before feature extraction.



   .. py:method:: forward_once(x)

      Encode one image batch into sorted SWD feature representation.



   .. py:method:: forward(x, y)

      Compute MS-SWD distance.

      :param x: Distorted image tensor with shape ``(N, 3, H, W)``.
      :type x: torch.Tensor
      :param y: Reference image tensor with shape ``(N, 3, H, W)``.
      :type y: torch.Tensor

      :returns: Distance scores with shape ``(N,)``.
      :rtype: torch.Tensor