Provided are a device and method for analyzing spatiotemporal data of a geographical space including one or more regions. The method includes imaging spatiotemporally varying data, selecting a representative image according to each transition state based on a spatiotemporal change in the imaged data, grouping pixels in the selected image by clustering the pixels, and performing group-specific data analyses.

Techniques for classifying biomedical image data using a graph neural network are disclosed. In one particular embodiment, the techniques may be realized as a method for classifying biomedical image data comprising generating an annotated representation of biomedical image data; identifying a plurality of pixel clusters based on the biomedical image data; constructing a graph based on the plurality of pixel clusters; determining at least one biomedical feature for at least one node of the graph based on the annotated representation of the biomedical image data; and processing a graph neural network to classify the biomedical image data based on the at least one biomedical feature.

A method performed by a computing device for clustering an image according to an embodiment of the present disclosure includes performing a first clustering on feature vectors of the plurality of images, and performing a second clustering for feature vectors belonging to some clusters that do not satisfy a reference score among clusters formed as a result of the first clustering, wherein a clustering parameter of the second clustering and a clustering parameter of the first clustering are different from each other.

Methods and systems for graph-based panoptic segmentation of point clouds are described herein. The methods receive points of a point cloud with a semantic label from a first category. Further, a plurality of unified cluster feature vectors from a second category are received. Each unified cluster feature vector is extracted from a cluster of points in the point cloud. A graph comprising nodes and edges is constructed from the plurality of unified cluster feature vectors. Each node of the graph is the unified feature vector, and each edge of the graph indicates the relationship between every two nodes of the graph. The edges of the graph are represented as an adjacency matrix, wherein the adjacency matrix indicates the existence, or the lack of existence, of an edge between every two nodes. The graph is fed to a graph convolutional neural network configured for predicting an instance label for each node or an attribute for each edge, wherein the attribute of each edge is used for assigning the instance label to each node. The method combines points with semantic labels for the first category and points with instance labels for the second category to generate points with both a sematic label and an instance label.

Provided herein are systems and methods for automatically segmenting a 3D model of a patient's teeth. A patient's dentition may be scanned. The scan data may be converted into a 3D model, including a graph-based representation of the 3D model. The graph-based representation can be input into a machine learning model to train the machine learning model to segment the 3D model into individual dental components. Trained machine learning models can also be used to segment graph-based representations of a 3D model of a patient's teeth.

Embodiments of the present disclosure describe mechanisms for a radar-based indoor localization and tracking system. One example can include monitoring unit that includes a radar source, a camera unit, and one or more processors coupled to the radar element and the camera unit. The monitoring unit is configured to generate point cloud data associated with an object; execute Point Cloud Library (PCL) preprocessing based, at least, on the point cloud data; execute Density-Based Spatial Clustering of Applications with Noise (DBSCAN) clustering; execute multi-object tracking on the object; and execute an image PCL overlay based on the point cloud data to generate real-time data associated with the object.

A computer-implemented method of characterizing a person's hand geometry includes inputting a three-dimensional (3D) point cloud of the person's hand into a clustered dynamic graph convolutional neural network (clustered DGCNN), and processing the 3D point cloud, with a shared network portion of the clustered DGCNN, to create a processed version of the three-dimensional point cloud. The method further includes, with a shape regression network portion of the clustered DGCNN, assigning each respective feature point in the processed version of the 3D point cloud to a corresponding one of a plurality of pre-defined clusters, and applying one or more transformations to the feature points assigned to each respective cluster to produce per cluster shape parameters that represent shapes associated with portions of the person's hand that correspond to associated ones of the pre-defined clusters. Each pre-defined cluster corresponds to a unique part of a hand's surface.

Embodiments of the present application provide a lane line processing method and a lane line processing device. The method can include: performing a binarization processing on a first image to obtain a binary image, the first image including lane line points and non-lane line points; performing a connected domain analysis on the binary image to obtain at least one connected domain in the binary image, the connected domain including a plurality of adjacent lane line points; determining lane line points in a group corresponding to a lane line, based on the connected domain; and obtaining representation information of the lane line corresponding to the group, by using the lane line points in the group.

This application discloses a method and apparatus for segmenting point cloud data, a storage medium and an electronic device. The method includes: obtaining target point cloud data by scanning target objects around a vehicle with laser lines; clustering the target point cloud data to obtain a plurality of first datasets, wherein feature points represented by point cloud data within each of the plurality of first datasets are fitted on one segmented line segment, each feature point being a point on a respective target object; and combining the plurality of first datasets according to distances between the corresponding plurality of segmented line segments to obtain a plurality of second datasets, wherein each second dataset includes at least one of the plurality of first datasets. This application resolves a technical problem of relatively low efficiency of point cloud segmentation in the related art.

A method includes accessing an audiovisual composition comprising a target video segment and a source video segment. The method also includes, in response to presence of the target video segment and the source video segment in the audiovisual composition: accessing a first keyword associated with the source video segment; and calculating a first relevance score for the first keyword relative to the target video segment based on a temporal position of the source video segment in the audiovisual composition and a temporal position of the target video segment in the audiovisual composition; accessing a textual query comprising the first keyword. The method additionally includes: generating a first query result based on the textual query, the first query result comprising the target video segment based on the first relevance score; and at a native composition application, rendering a representation of the first query result.