Same as to in torch module Don't really underestand why this isn't a module in the first place

Create a 1D, 2D, or 3D convolution module.

3x3 convolution with padding

one dimensional cubic interpolation :param p: [N, 4] (4) should be in order :param x: [N] :return:

:param feature_maps: pyramid features (B,V,C0+C1+C2,H,W) fused pyramid features :param partial_vol_origin: [B, 3] the world coordin

get feature maps of all conditional images :param imgs: :return:

sample features of pts from 2d feature maps :param pts: [N_rays, N_samples, 3] :param featureMaps: [N_views, C, H, W] :param w2cs: [N

Make a standard normalization layer. :param channels: number of input channels. :return: an nn.Module for normalization.

1x1 convolution without padding

generate grid if 3D volume, grid[:,:,x,y,z] = (x,y,z) :param n_vox: :param interval: :return:

img_h: int, an image height img_w: int, an image width marg: int, a margin for a box starting coordinate p_irr: float, 0 <= p_irr <=

:param pts: [N, 3] :param mask_volume: [1, 1, X, Y, Z] :return:

Create a linear module.

Imagenet Superresolution Dataloader Performs following ops in order: 1. crops a crop of size s from image either as random o

two dimensional cubic interpolation :param p: [N, 4, 4] :param x: [N] :param y: [N] :return:

:param polar_deg (float). :param azimuth_deg (float). :param radius_m (float). :param fov_deg (float). :return (5, 3) array of fl

:param conditional_volume: [1,C, dX,dY,dZ] :param mask_volume: [1,1, dX,dY,dZ] :param coords_volume: [1,3, dX,dY,dZ]

assume batch size == 1, from the first lod to get sparse voxels :param sdf_volume: [num_pts, 1] :param coords_volume: [3, X,

assume batch size == 1, from the first lod to get sparse voxels :param sdf_volume: [1, X, Y, Z] :param coords_volume: [3, X,

bilinear sampling cannot guarantee continuous first-order gradient mimic pytorch grid_sample function The 8 corner points of a volume not

replace background pixel with random color in rendering

sample feature of pts_wrd from featureVolume, all in world space :param pts: [N_rays, n_samples, 3] :param featureVolume: [C,wX,wY,wZ]