open3d.ml.torch.ops

Functional API with operators.

These are the building blocks for the layers. See The layer API for an easy to use high level interface.

Functions

build_spatial_hash_table(points, radius, ...)	Creates a spatial hash table meant as input for fixed_radius_search
continuous_conv(filters, out_positions, ...)	Continuous convolution of two pointclouds.
continuous_conv_transpose(filters, ...[, ...])	Continuous tranpose convolution of two pointclouds.
fixed_radius_search(points, queries, radius, ...)	Computes the indices of all neighbors within a radius.
invert_neighbors_list(num_points, ...)	Inverts a neighbors list made of neighbors_index and neighbors_row_splits.
knn_search(points, queries, k, ...[, ...])	Computes the indices of k nearest neighbors.
nms(boxes, scores, nms_overlap_thresh)	Performs non-maximum suppression of bounding boxes.
radius_search(points, queries, radii, ...[, ...])	Computes the indices and distances of all neighbours within a radius.
reduce_subarrays_sum(values, row_splits)	Computes the sum for each subarray in a flat vector of arrays.
voxel_pooling(positions, features, voxel_size)	Spatial pooling for point clouds by combining points that fall into the same voxel bin.
voxelize(points, row_splits, voxel_size, ...)	Voxelization for point clouds.