pyiqa.archs.nrqm_arch
=====================

.. py:module:: pyiqa.archs.nrqm_arch

.. autoapi-nested-parse::

   NRQM Metric, proposed in

   Chao Ma, Chih-Yuan Yang, Xiaokang Yang, Ming-Hsuan Yang
   "Learning a No-Reference Quality Metric for Single-Image Super-Resolution"
   Computer Vision and Image Understanding (CVIU), 2017

   Matlab reference: https://github.com/chaoma99/sr-metric
   This PyTorch implementation by: Chaofeng Chen (https://github.com/chaofengc)









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

.. py:data:: default_model_urls

.. py:function:: get_gauss_pyramid(x: torch.Tensor, scale: int = 2)

   Get gaussian pyramid images with gaussian kernel.


.. py:function:: get_var_gen_gauss(x, eps=1e-07)

   Get mean and variance of input local patch.


.. py:function:: gamma_gen_gauss(x: torch.Tensor, block_seg=10000.0)

   General gaussian distribution estimation.

   :param block_seg: maximum number of blocks in parallel to avoid OOM


.. py:function:: gamma_dct(dct_img_block: torch.Tensor)

   Generalized gaussian distribution features


.. py:function:: coeff_var_dct(dct_img_block: torch.Tensor)

   Gaussian var, mean features


.. py:function:: oriented_dct_rho(dct_img_block: torch.Tensor)

   Oriented frequency features


.. py:function:: block_dct(img: torch.Tensor)

   Get local frequency features


.. py:function:: norm_sender_normalized(pyr, num_scale=2, num_bands=6, blksz=3, eps=1e-12)

   Normalize pyramid with local spatial neighbor and band neighbor


.. py:function:: global_gsm(img: torch.Tensor)

   Global feature from gassian scale mixture model


.. py:function:: tree_regression(feat, ldau, rdau, threshold_value, pred_value, best_attri)

   Simple decision tree regression.


.. py:function:: random_forest_regression(feat, ldau, rdau, threshold_value, pred_value, best_attri)

   Simple random forest regression.

   Note: currently, this is non-differentiable and only support CPU.


.. py:function:: nrqm(img: torch.Tensor, linear_param, rf_param) -> torch.Tensor

   Calculate NRQM
   :param img: Input image.
   :type img: Tensor
   :param linear_param: (4, 1) linear regression params
   :type linear_param: np.array
   :param rf_param: params of 3 random forest for 3 kinds of features


.. py:function:: calculate_nrqm(img: torch.Tensor, crop_border: int = 0, test_y_channel: bool = True, color_space: str = 'yiq', linear_param: torch.Tensor = None, rf_params_list: list = None, **kwargs) -> torch.Tensor

   Calculate NRQM
   :param img: Input image whose quality needs to be computed.
   :type img: Tensor
   :param crop_border: Cropped pixels in each edge of an image. These
                       pixels are not involved in the metric calculation.
   :type crop_border: int
   :param test_y_channel: Whether converted to 'y' (of MATLAB YCbCr) or 'gray'.
   :type test_y_channel: Bool
   :param pretrained_model_path: The pretrained model path.
   :type pretrained_model_path: String

   :returns: NIQE result.
   :rtype: Tensor


.. py:class:: NRQM(test_y_channel: bool = True, color_space: str = 'yiq', crop_border: int = 0, pretrained_model_path: str = None)

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


   NRQM metric

   Ma, Chao, Chih-Yuan Yang, Xiaokang Yang, and Ming-Hsuan Yang.
   "Learning a no-reference quality metric for single-image super-resolution."
   Computer Vision and Image Understanding 158 (2017): 1-16.

   :param - channels: Number of processed channel.
   :type - channels: int
   :param - test_y_channel: whether to use y channel on ycbcr.
   :type - test_y_channel: Boolean
   :param - crop_border: Cropped pixels in each edge of an image. These
                         pixels are not involved in the metric calculation.
   :type - crop_border: int
   :param - pretrained_model_path: The pretrained model path.
   :type - pretrained_model_path: String


   .. py:method:: forward(X: torch.Tensor) -> torch.Tensor

      Computation of NRQM metric.
      :param X: An input tensor. Shape :math:`(N, C, H, W)`.

      :returns: Value of nrqm metric.



.. py:class:: PI(crop_border=4, color_space='ycbcr')

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


   Perceptual Index (PI), introduced by

   Blau, Yochai, Roey Mechrez, Radu Timofte, Tomer Michaeli, and Lihi Zelnik-Manor.
   "The 2018 pirm challenge on perceptual image super-resolution."
   In Proceedings of the European Conference on Computer Vision (ECCV) Workshops, pp. 0-0. 2018.
   Ref url: https://github.com/roimehrez/PIRM2018

   It is a combination of NIQE and NRQM: 1/2 * ((10 - NRQM) + NIQE)

   :param - color_space: color space of y channel, default ycbcr.
   :type - color_space: str
   :param - crop_border: Cropped pixels in each edge of an image, default 4.
   :type - crop_border: int


   .. py:method:: forward(X: torch.Tensor) -> torch.Tensor

      Computation of PI metric.
      :param X: An input tensor. Shape :math:`(N, C, H, W)`.

      :returns: Value of PI metric.