open3d.core.nns.NearestNeighborSearch

class open3d.core.nns.NearestNeighborSearch

NearestNeighborSearch class for nearest neighbor search. Construct a NearestNeighborSearch object with input dataset_points of shape {n_dataset, d}.

__init__(self: open3d.cpu.pybind.core.nns.NearestNeighborSearch, dataset_points: open3d.cpu.pybind.core.Tensor, index_dtype: open3d.cpu.pybind.core.Dtype = Int64) → None

fixed_radius_index(self: open3d.cpu.pybind.core.nns.NearestNeighborSearch, radius: float | None = None) → bool

fixed_radius_search(self, query_points, radius, sort=None)

Parameters:

• query_points (open3d.core.Tensor) – The query tensor of shape {n_query, d}.

• radius (float) – Radius value for radius search.

• sort (Optional[bool], optional, default=None) –

Returns:

tuple[open3d.core.Tensor, open3d.core.Tensor, open3d.core.Tensor]

hybrid_index(self: open3d.cpu.pybind.core.nns.NearestNeighborSearch, radius: float | None = None) → bool

hybrid_search(self, query_points, radius, max_knn)

Perform hybrid search.

Parameters:

• query_points (open3d.core.Tensor) – The query tensor of shape {n_query, d}.

• radius (float) – Radius value for radius search.

• max_knn (int) – Maximum number of neighbors to search per query point.

Returns:

tuple[open3d.core.Tensor, open3d.core.Tensor, open3d.core.Tensor]

knn_index(self: open3d.cpu.pybind.core.nns.NearestNeighborSearch) → bool

Set index for knn search.

knn_search(self, query_points, knn)

Perform knn search.

Parameters:

• query_points (open3d.core.Tensor) – The query tensor of shape {n_query, d}.

• knn (int) – Number of neighbors to search per query point.

Returns:

tuple[open3d.core.Tensor, open3d.core.Tensor]

multi_radius_index(self: open3d.cpu.pybind.core.nns.NearestNeighborSearch) → bool

Set index for multi-radius search.

multi_radius_search(self, query_points, radii)

Perform multi-radius search. Each query point has an independent radius.

Parameters:

• query_points (open3d.core.Tensor) – The query tensor of shape {n_query, d}.

• radii (open3d.core.Tensor) – Tensor of shape {n_query,} containing multiple radii, one for each query point.

Returns:

tuple[open3d.core.Tensor, open3d.core.Tensor, open3d.core.Tensor]