cvtkit.common
A suite of common utility functions.
This module includes general utility functions used by the sub-packages of the CVTkit library.
build_coords_list(H, W, batch_size, device)
Constructs an batched index list of pixel coordinates.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
H
|
int
|
Height of the pixel grid. |
required |
W
|
int
|
Width of the pixel grid. |
required |
batch_size
|
int
|
Number of batches. |
required |
device
|
str
|
GPU device identifier. |
required |
Returns:
| Type | Description |
|---|---|
Tensor
|
The index list of shape [batch_size, H*W, 2] |
build_depth_pyramid(depth, levels)
build_labels(depth, hypotheses)
cosine_similarity(t1, t2)
Computes the cosine similarity between two tensors.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
t1
|
Tensor
|
First tensor. |
required |
t2
|
Tensor
|
Second tensor. |
required |
Returns:
| Type | Description |
|---|---|
Tensor
|
The cosine similarity between the two tensors. |
crop_image(image, crop_row, crop_col, scale)
freeze_model_weights(model)
gaussian(window_size, sigma)
groupwise_correlation(t1, t2, num_groups)
Computes the Group-Wise Correlation (GWC) between two tensors.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
t1
|
Tensor
|
First tensor. |
required |
t2
|
Tensor
|
Second tensor. |
required |
num_groups
|
int
|
Number of groups. |
required |
Returns:
| Type | Description |
|---|---|
Tensor
|
The Group-Wise Correlation (GWC) between the two tensors. |
laplacian_pyramid(image)
Computes the Laplacian pyramid of an image.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
image
|
Tensor
|
2D map to compute Laplacian over. |
required |
tau
|
Laplacian region threshold. |
required |
Returns:
| Type | Description |
|---|---|
Tensor
|
The map of the Laplacian regions. |
laplacian_pyramid_th(image, tau)
Computes the Laplacian pyramid of an image.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
image
|
Tensor
|
2D map to compute Laplacian over. |
required |
tau
|
float
|
Laplacian region threshold. |
required |
Returns:
| Type | Description |
|---|---|
Tensor
|
The map of the Laplacian regions. |
non_zero_std(maps, device, dim=1, keepdim=False)
Computes the standard deviation of all non-zero values in an input Tensor along the given dimension.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
maps
|
Tensor
|
|
required |
device
|
str
|
|
required |
keepdim
|
bool
|
|
False
|
Returns:
| Type | Description |
|---|---|
Tensor
|
The standard deviation of the non-zero elements of the input map. |
normalize(data, min_val=None, max_val=None)
parameters_count(net, name, do_print=True)
Parameters:
Returns:
print_gpu_mem()
Prints the current unallocated memory of the GPU.
round_nearest(num, decimal=0)
Rounds a floating point number to the nearest decimal place.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
num
|
float
|
Float to be rounded. |
required |
decimal
|
int
|
Decimal place to round to. |
0
|
Returns:
| Type | Description |
|---|---|
int
|
The given number rounded to the nearest decimal place. |
Examples:
>>> round_nearest(11.1)
11
>>> round_nearest(15.7)
16
>>> round_nearest(2.5)
2
>>> round_nearest(3.5)
3
>>> round_nearest(14.156, 1)
14.2
>>> round_nearest(15.156, 1)
15.2
>>> round_nearest(15.156, 2)
15.16
scale_camera(cam, scale=1.0)
Scales a camera intrinsic parameters.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
cam
|
ndarray
|
Input camera to be scaled. |
required |
scale
|
float
|
Scale factor. |
1.0
|
Returns:
| Type | Description |
|---|---|
ndarray
|
The scaled camera. |
scale_image(image, scale=1.0, interpolation='linear')
Scales an input pixel grid.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
image
|
ndarray
|
Input image to be scaled. |
required |
scale
|
float
|
Scale factor. |
1.0
|
interpolation
|
str
|
Interpolation technique to be used. |
'linear'
|
Returns:
| Type | Description |
|---|---|
NDArray[float32]
|
The scaled image. |
scale_mvs_data(depths, confs, cams, scale=1.0, interpolation='linear')
Scales input depth maps, confidence maps, and cameras.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
depths
|
ndarray
|
Input depth maps to be scaled. |
required |
confs
|
ndarray
|
Input confidence maps to be scaled. |
required |
cams
|
ndarray
|
Input cameras to be scaled |
required |
scale
|
float
|
Scale factor. |
1.0
|
interpolation
|
str
|
Interpolation technique. |
'linear'
|
Returns:
| Name | Type | Description |
|---|---|---|
scaled_depths |
ndarray
|
The scaled depth maps. |
scaled_confs |
ndarray
|
The scaled confidence maps. |
cams |
ndarray
|
The scaled cameras. |
set_random_seed(seed)
Parameters:
Returns:
to_gpu(data, device)
Loads a dictionary of elements onto the GPU device.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
data
|
dict
|
Dictionary to be loaded. |
required |
device
|
str
|
GPU device identifier. |
required |
y_axis_rotation(P, theta)
Applies a rotation to the given camera extrinsics matrix along the y-axis.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
P
|
ndarray
|
Initial extrinsics camera matrix. |
required |
theta
|
float
|
Angle (in radians) to rotate the camera. |
required |
Returns:
| Type | Description |
|---|---|
ndarray
|
The rotated extrinsics matrix for the camera. |