Returns the dim'th component of the idx'th point in the class:

Read and unpack the next bytes from a binary file. :param fid: :param num_bytes: Sum of combination of {2, 4, 8}, e.g. 2, 6, 16, 30, etc.

Normalization helper function.

Helper accessor to the dataset points:

Transforms inputs between different color spaces :param input_color: tensor of colors to transform (with CxHxW layout) :param fromSpace2toSpace:

Must return the number of data points

Computes a weighted percentile of an error map, i.e., the error which is such that the sum of the errors in the error map that are smaller than it

r""" Blur input with 1-D kernel Args: input (torch.Tensor): a batch of tensors to be blured window (torch.Tensor): 1-D gauss kerne

Saves image as png :param img_file: image's filename :param img: float image (in the [0,1] range) to save

Transforms an image from CxHxW layout to HxWxC :param x: image with CxHxW layout :return: image with HxWxC layout

r"""Create 1-D gauss kernel Args: size (int): the size of gauss kernel sigma (float): sigma of normal distribution Returns:

Normalization helper function.

Detects edges and points (features) in the achromatic image :param imgy: achromatic image (on 1xHxW layout, containing normalized Y-values from YC

Applies Hunt-adjustment to an image :param img: image to adjust (with CxHxW layout in the L*a*b* color space) :return: Hunt-adjusted image (with

Apply `fn` to each element of `tup` and cast to `tup`'s namedtuple.

Peek at the next training batch or test example without dequeuing it. Returns: batch: dict, has "pixels" and "rays".

Transforms an image from HxWxC layout to CxHxW :param x: image with HxWxC layout :return: image with CxHxW layout

Not a leaf, but a internal node

* Builds the index */

Frees the previously-built index. Automatically called within * buildIndex(). */

Generates spatial contrast sensitivity filters with width depending on the number of pixels per degree of visual angle of the observer :param pix

[H, W, C], [0, 1]

[H, W, C], [0, 1]

Rebuild the KD tree from scratch. Call if data updated.

Push node and branch onto iteration stack (For internal use only)

Generating rays for all images.

Load images from disk.

Average poses according to the original NeRF code.

r""" Calculate ssim index for each color channel of X and Y Args: X (torch.Tensor): images Y (torch.Tensor): images win (t

Construct lookat view matrix.

* Called during search to add an element matching the criteria. * @return true if the search should be continued, false if the results are * suf

Checks reference and test images for NaNs and sets NaNs to 0. Depending on verbosity level, warns if NaNs occur :param reference: float tensor :p

[B, C, H, W], [0, 1]

[H, W, C], [0, 1]

Computes the FLIP error map between two LDR images, assuming the images are observed at a certain number of pixels per degree of visual angle :p

[B, C, H, W], [0, 1]

[B, C, H, W], [0, 1]

* Create a tree node that subdivides the list of vecs from vind[first] * to vind[last]. The routine is called recursively on each sublist. *

Return the magma map, described here: https://bids.github.io/colormap/ :return: matrix of size 256x3 where each row contains a color in the magma

Return a boolean indicating whether a flag name is supported on the specified compiler.

Computes the HyAB distance between reference and test images :param reference: reference image (with CxHxW layout in the standard or Hunt-adjusted

Transforms grayscale index map to colors in a color map :param index_map: integer matrix containing integer versions of the FLIP errors, scaled to

Lift a Gaussian defined along a curved ray to 3D coordinates.

Loads a previous index from a binary file. * IMPORTANT NOTE: The set of data points is NOT stored in the file, so the * index object must be c

Loads an image (in the [0, 255] range) and transforms it into a numpy array and into the [0, 1] range :param img_file: image's filename :return:

Args - pts: np.ndarray, [H, W, D, 4] - cam_mat: np.ndarray, [4, 4] - view_mat: np.ndarray, [4, 4] Returns: - uv: np.ndarray, [H,

Read color data from EXR image file. Set negative values and nans to 0. :param filename: string describing file path :return: RGB image in float3

Stores the index in a binary file. * IMPORTANT NOTE: The set of data points is NOT stored in the file, so when * loading the index object it m

Filters an image with channel specific spatial contrast sensitivity functions and clips result to the unit cube in linear RGB :param img: image t

Args: - mat: np.ndarray, [4, 4] Returns: - ret: np.ndarray, [4, 4]

Applies exposure compensation and tone mapping. Refer to the Visualizing Errors in Rendered High Dynamic Range Images paper for details about the

Visualize a depth map. Args: depth: A depth map. acc: An accumulation map, in [0, 1]. near: The depth of the near plane, if None then j

Reset iterator