RHM.geometry_utils

Functions

compute_embedding(mesh1[, n_samples, ...])	Compute embedding of a mesh using MDS
decimate(mesh1, n_target_faces[, n_jobs])	Runs quadratic mesh decimation on a mesh
knn_query(X, Y[, k, return_distance, n_jobs])	Query nearest neighbors.

RHM.geometry_utils.knn_query(X, Y, k=1, return_distance=False, n_jobs=1)

Query nearest neighbors.

Parameters:

• X ((n,d) np.ndarray) – Dataset

• Y ((m,d) np.ndarray) – Query points

• k (int) – Number of neighbors to return

• return_distance (bool) – Whether to return distances

• n_jobs (int) – Number of jobs to run in parallel

Returns:

• dists ((m,k) np.ndarray, optional) – Distances to the k nearest neighbors. Only returned if return_distance is True

• matches ((m,k) np.ndarray) – Indices of the k nearest neighbors

RHM.geometry_utils.decimate(mesh1, n_target_faces, n_jobs=1)

Runs quadratic mesh decimation on a mesh

Parameters:

• mesh1 (TriMesh object) – input mesh

• n_target_faces (int) – number of faces in the output mesh

• n_jobs (int) – number of jobs to run in parallel for NN query

Returns:

sub_indices – Indices of the vertices in the decimated mesh

Return type:

(m,) np.ndarray

RHM.geometry_utils.compute_embedding(mesh1, n_samples=500, n_components=8, n_jobs=1)

Compute embedding of a mesh using MDS

The embedding mimics the geodesic distances of the mesh

Parameters:

• mesh1 (TriMesh object) – input mesh

• n_samples (int) – number of samples to use for the embedding

• n_components (int) – number of components of the embedding

• n_jobs (int) – number of jobs to run in parallel for MDS

Returns:

emb_final – Embedding of the mesh

Return type:

(n,d) np.ndarray