pyiqa.archs.pieapp_arch
=======================

.. py:module:: pyiqa.archs.pieapp_arch

.. autoapi-nested-parse::

   PieAPP metric, proposed by

   Prashnani, Ekta, Hong Cai, Yasamin Mostofi, and Pradeep Sen.
   "Pieapp: Perceptual image-error assessment through pairwise preference."
   In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1808-1817. 2018.

   Ref url: https://github.com/prashnani/PerceptualImageError
   Modified by: Chaofeng Chen (https://github.com/chaofengc)

   !!! Important Note: to keep simple test process and fair comparison with other methods,
                       we use zero padding and extract subpatches only once
                       rather than from multiple subimages as the original codes.







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

.. py:data:: default_model_urls

.. py:class:: CompactLinear

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


   .. py:method:: forward(x)


.. py:class:: PieAPP(patch_size=64, stride=27, pretrained=True, pretrained_model_path=None)

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


   PieAPP model implementation.

   :param - patch_size: Size of the patches to extract from the images.
   :type - patch_size: int
   :param - stride: Stride to use when extracting patches.
   :type - stride: int
   :param - pretrained: Whether to use a pretrained model or not.
   :type - pretrained: bool
   :param - pretrained_model_path: Path to the pretrained model.
   :type - pretrained_model_path: str

   .. method:: - flatten(matrix)

      Takes NxCxHxW input and outputs NxHWC.

   .. method:: compute_features(input)

      Computes the features of the input image.

   .. method:: - preprocess(x)

      Preprocesses the input image.

   .. method:: forward(dist, ref)

      Computes the PieAPP score between the distorted and reference images.
      


   .. py:method:: flatten(matrix)


   .. py:method:: compute_features(input)


   .. py:method:: preprocess(x)

      Default BGR in [0, 255] in original codes



   .. py:method:: forward(dist, ref)