Pure Python NMS baseline.

Calculate factor corresponding to quality Args: quality(float): Quality for jpeg compression. Returns: float: Compres

Randomly generate bivariate isotropic or anisotropic Gaussian kernels. In the isotropic mode, only `sigma_x_range` is used. `sigma_y_range` and

Randomly generate bivariate generalized Gaussian kernels. In the isotropic mode, only `sigma_x_range` is used. `sigma_y_range` and `rotation_ra

Randomly generate bivariate plateau kernels. In the isotropic mode, only `sigma_x_range` is used. `sigma_y_range` and `rotation_range` is ignor

Randomly generate mixed kernels. Args: kernel_list (tuple): a list name of kernel types, support ['iso', 'aniso', 'skew',

Save networks and training state.

Get human readable file size. Args: size (int): File size. suffix (str): Suffix. Default: 'B'. Return: str: F

Args: x: (b, h, w, c) window_size (int): window size Returns: windows: (num_windows*b, window_size, window_siz

It is the training pair pool for increasing the diversity in a batch. Batch processing limits the diversity of synthetic degradations in a

It is the training pair pool for increasing the diversity in a batch. Batch processing limits the diversity of synthetic degradations in a

Take the prediction of fake and real images from the combined batch. The prediction contains the intermediate outputs of multiscale

Return an activation function given a string

Get the initial lr, which is set by the scheduler.

Initialize slurm distributed training environment. If argument ``port`` is not specified, then the master port will be system environment

Postprocessing of raw pytorch model output. Returns: bboxes: list of arrays with 4 coordinates of bounding boxes with

Preprocessing image before passing through the network. Resize and conversion to torch tensor.

Print keys with different name or different size when loading models. 1. Print keys with different names. 2. If strict=False, prin

Set learning rate for warm-up. Args: lr_groups_l (list): List for lr_groups, each for an optimizer.

Calculate SSIM (structural similarity) for one channel images. It is called by func:`calculate_ssim`. Args: img (ndarray): Imag

Calculate SSIM (structural similarity) (PyTorch version). It is called by func:`calculate_ssim_pt`. Args: img (Tensor): Images

Adaptive instance normalization. Adjust the reference features to have the similar color and illuminations as those in the degradate featu

Adaptive instance normalization. Adjust the reference features to have the similar color and illuminations as those in the degradate featu

Add JPG compression artifacts. Args: img (Numpy array): Input image, shape (h, w, c), range [0, 1], float32. quality (float):

Align and crop face with landmarks. The output_size and transform_size are based on width. The height is adjusted based on shrink_ratio_h/

Align and warp faces with face template.

Decode locations from predictions using priors to undo the encoding we did for offset regression at train time. Args: b_loc (tensor

Decode landm from predictions using priors to undo the encoding we did for offset regression at train time. Args: pre (tensor): lan

Arguments: frames: a list of PIL.Image, or torch.Tensor(shape=[n, h, w, c], type=np.float32, BGR format).

Generate a bivariate isotropic or anisotropic Gaussian kernel. In the isotropic mode, only `sig_x` is used. `sig_y` and `theta` is ignored.

Generate a bivariate generalized Gaussian kernel. Described in `Parameter Estimation For Multivariate Generalized Gaussian Distribut

Generate a plateau-like anisotropic kernel. 1 / (1+x^(beta)) Ref: https://stats.stackexchange.com/questions/203629/is-there-a-plateau-shap

Check whether the input is a mirror-extended sequence. If mirror-extended, the i-th (i=0, ..., t-1) frame is equal to the (t-1-i)-th

Check resume states and pretrain_network paths. Args: opt (dict): Options. resume_iter (int): Resume iteration.

Compute features. Args: block (ndarray): 2D Image block. Returns: list: Features with length of 18.

Compute optical flow using SPyNet for feature alignment. Note that if the input is an mirror-extended sequence, 'flows_forward' is n

Conv block used in MSDilationBlock.

Function: ---------- Convert Transform Matrix 'trans' into 'cv2_trans' which could be directly used by cv2.warpAffine():

Decode locations from predictions using priors to undo the encoding we did for offset regression at train time. Args: loc (tensor):

Decode landm from predictions using priors to undo the encoding we did for offset regression at train time. Args: pre (tensor): lan