KMeans with both random and KMeans++ initialization Parameters ---------- n_clusters: int Number of clusters tolerance:

KITTI point cloud classification dataset for deep learning object classification Parameters ---------- input_dir: str Direc

KDTree for kNN & RNN search Parameters ---------- point_cloud : numpy.ndarray Point cloud measurements as N-by-3 numpy.ndarr

ModelNet40 dataset for deep learning based point cloud classification Parameters ---------- input_dir: str Directory path o

OCTree for kNN & RNN search Parameters ---------- point_cloud : numpy.ndarray Point cloud measurements as N-by-3 numpy.ndarr

SpectralClustering Parameters ---------- n_clusters: int Number of clusters Attributes ----------

WebSockets server class. As an alternative, the standard library SocketServer can be used

Verifies Basic Auth headers. Specify src as username:password

Ground segmenter Parameters ---------- cloud: pcl.cloud PCL point cloud max_distance: float RANSAC inlier thresh

Just like WebSocketProxy, but uses standard Python SocketServer framework.

WebSocket protocol socket like class. This provides access to the WebSocket protocol by behaving much like a real socket would. It shares man

Proxy traffic to and from a WebSockets client to a normal TCP socket server target.

DBSCAN segmenter for surrounding objects Parameters ---------- cloud: pcl.cloud PCL point cloud eps: float RANSA

Node in KDTree Parameters ---------- axis : int Splitting axis value: int Splitting

Octant in OCTree Parameters ---------- axis : int Splitting axis value: int Splittin

Outlier filter Parameters ---------- cloud: pcl.cloud PCL point cloud k: int the number of nearest neighbor poin

ROI filter Parameters ---------- cloud: pcl.cloud PCL point cloud name: string filter field name limits: tup