initialize network's weights init_type: normal | xavier | kaiming | orthogonal https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix/bl

Calculates the gradient penalty loss for WGAN GP

3x3 convolution with padding

r"""Function that blurs a tensor using a Gaussian filter. See :class:`~kornia.filters.GaussianBlur` for details.

r"""Function that returns Gaussian filter coefficients. Args: kernel_size (int): filter size. It should be odd and positive. sigm

Predict the landmarks for each face present in the image. This function predicts a set of 68 2D or 3D images, one for each image present.

deploy for nsh

Use a slow but more flexible looping method to process face lines as a fallback option to faster vectorized methods.

Parse loaded homogeneous (tri/quad) face data in a vectorized manner.

Computes zero padding tuple.

r"""Function that returns Gaussian filter matrix coefficients. Args: kernel_size (Tuple[int, int]): filter sizes in the x and y direction

Parse raw OBJ text into vertices, vertex normals, vertex colors, and vertex textures.

Constructs a ResNet-50 model. Args: pretrained (bool): If True, returns a model pre-trained on ImageNet

Get the color in a texture image. Parameters ------------- uv : (n, 2) float UV coordinates on texture image image : PIL.Image Tex

Save 3D face model with texture represented by colors. Args: obj_name: str vertices: shape = (nver, 3) triangles: shape =

Save 3D face model with texture represented by colors. Args: obj_name: str vertices: shape = (nver, 3) triangles: shape =

if game_lm is not None, the result (mesh obj and UV texture map) will be convert from nsh to the game