Removes keypoints too close to the border

Fast Non-maximum suppression to remove nearby points

Plot lines for existing images. Args: img_path: string, path to the image. lines2D: list of ndarrays of size (N, 4). line

Plot keypoints for existing images. Args: img_path: string, path to the image. points2D: list of ndarrays of size (N, 2).

Plot keypoints for existing images. Args: img_path: string, path to the image. points2D: list of ndarrays of size (N, 2).

see: src/base/reconstruction.cc void Reconstruction::WriteCamerasBinary(const std::string& path) void Reconstruction::ReadCameras

see: src/base/reconstruction.cc void Reconstruction::WriteCamerasText(const std::string& path) void Reconstruction::ReadCamerasTe

see: src/base/reconstruction.cc void Reconstruction::ReadImagesBinary(const std::string& path) void Reconstruction::WriteImagesBi

see: src/base/reconstruction.cc void Reconstruction::ReadImagesText(const std::string& path) void Reconstruction::WriteImagesText

see: src/base/reconstruction.cc void Reconstruction::ReadPoints3DBinary(const std::string& path) void Reconstruction::WritePoints

see: src/base/reconstruction.cc void Reconstruction::ReadPoints3DText(const std::string& path) void Reconstruction::WritePoints3D

corresponding to show following losses: ['total_loss', 'point_loss', 'point_uncer_loss', 'line_loss',

- image class for storing 2D & 3D points, 2D & 3D lines, camera vector, pose vector - comments with ### can be changed if augmenting

highlight_frame: "seq-06/frame-000612.color.png", for example limit_n_frames: limit the number of frames to visualize the 3D lines an

Compute dense scores and descriptors for an image

Correct the points and lines coordinates to be inside the image if the points are outside the image, remove them if lines have half insi

Apply the augmentation to the input image, points, lines, camera, and pose args: image: input image np.array WxH cfg: config

Check the data and call the _forward method of the child model.

get uncertainty and remove line with high uncertainty args: lines3D: numpy array (1, 7, N) return: lines3D (N, 6)

Return a list of points2D and a list of lines2D in the image

Return True if the augmentation is applied by select a random option

To be implemented by child class.

Normalize the image tensor and reorder the dimensions.

Visualize 3D lines from datasetcollection Args: datacollection: DataCollection object train_or_test: string, "train" or "test

Visualize 3D lines from datasetcollection Args: datacollection: DataCollection object img_name: string, image name

Visualize 3D points from datasetcollection Args: datacollection: DataCollection object mode: string, "train" or "test"

Sample a random valid homography. Computes the homography transformation between a random patch in the original image and a warped projection