To estimate the top surface of a measurement target on the basis of a data point group that corresponds to the top surface of a measurement target and is obtained using a laser scanner. This top surface estimation method for hoisting loads and by acquiring, using the laser scanner, data point groups in a hoisting load region which includes a hoisting load and an object from above the hoisting load and the object, dividing the hoisting load region into layers which constitute a plurality of groups which have a prescribed thickness in the vertical direction, and allocating the acquired data point groups to the plurality of layer groups, and estimating the top surfaces of the hoisting load and the object in each layer group on the basis of the data point groups allocated to the plurality of layer groups.

Physical object boundary detection techniques and systems are described. In one example, an augmented reality module generates three dimensional point cloud data. This data describes depths at respective points within a physical environment that includes the physical object. A physical object boundary detection module is then employed to filter the point cloud data by removing points that correspond to a ground plane. The module then performs a nearest neighbor search to locate a subset of the points within the filtered point cloud data that correspond to the physical object. Based on this subset, the module projects the subset of points onto the ground plane to generate a two-dimensional boundary. The two-dimensional boundary is then extruded based on a height determined from a point having a maximum distance from the ground plane from the filtered cloud point data.

Systems and methods for generating a digital model of a space and modifying the digital model are described. In one aspect, a system includes one or more computers and one or more storage devices storing instructions that, when executed by the one or more computers, cause the one or more computers to perform operations including: obtaining a point cloud representation of the space, the point cloud including multiple points; segmenting the point cloud into: (i) an inlier point cloud including multiple inlier points, and (ii) an outlier point cloud including multiple outlier points, where the segmenting includes: identifying, as the inlier points, the points of the point cloud that have at least a specified likelihood of being measurements of any of multiple planes of the space; and identifying, as the outlier points, all other points of the point cloud that are not identified as the inlier points; processing the inlier point cloud to generate a segmented inlier point cloud that includes, for each measured plane of the space, a respective plane point cloud representing the plane; and processing, using a neural network, the outlier point cloud to generate a segmented outlier point cloud that includes, for each of one or more objects detected in the space, a respective object point cloud representing the object.

The present invention relates to a method for arranging one or more functional elements in a room, comprising the steps of: a) three-dimensional acquisition of the room or of a part of the room by means of a scanner, b) definition of a grid based on the acquisition according to step a), c) definition of grid points of the room in the grid established according to step b), d) definition of at least one grid point of one or more functional elements, and e) virtual arrangement of the one or more functional elements in the room at at least one location, at which at least one grid point of the room and at least one grid point of the functional element coincide.

The disclosure provides a structure diagnosis system and a structure diagnosis method. The structure diagnosis system includes: a lidar scanner scanning a structure to generate a point cloud data; an input interface receiving the point cloud data; and a processor receiving the point cloud data and generating a point cloud data set. The processor executes a surface degradation and geometry abnormal coupling diagnosis module to: marking a first point cloud range of a surface degradation area according to color space value of the point cloud data set; marking a second point cloud range of a geometry abnormal area according to coordinate value of the point cloud data set; when an abnormal area includes the first point cloud range and the second point cloud range at least partially overlapping each other, determining surface degradation or geometry abnormal occurring at the abnormal area and mark the abnormal area with a predetermined mode.

An electronic apparatus performs a method of recovering a complete and dense point cloud from a partial point cloud. The method includes: constructing a sparse but complete point cloud from the partial point cloud through a contrastive teacher-student neural network; and transforming the sparse but complete point cloud to the complete and dense point cloud. In some embodiments, the contrastive teacher-student neural network has a dual network structure comprising a teacher network and a student network both sharing the same architecture. The teacher network is a point cloud self-reconstruction network, and the student network is a point cloud completion network.

Occupancy networks enable efficient and flexible point cloud compression. In addition to the voxel-based representation, occupancy networks are able to handle points, meshes, or projected images of 3D objects, making them very flexible in terms of input signal representation. The probability of occupancy of positions is estimated using occupancy networks instead of sparse convolutional neural networks. A compression implementation using occupancy network enables scalability with infinite reconstruction resolution.

A system and method for providing positional error correction for particles of destructible objects in a three-dimensional volume in a virtual space includes electronic storage to store center-of-mass information of a set of objects, using a high-precision floating point format. Prior to runtime and/or interactive manipulation of the set of objects, a texture map is generated that includes positional information in a floating point format having less precision than the high-precision floating point format. A simulation uses this texture map to determine simulated center-of-mass information of the set of objects. This simulated center-of-mass information is compared with the previously stored center-of-mass information to determine which objects have positional errors, and which offsets are needed to correct the positional errors. The stored center-of-mass information is adjusted by the determined offsets, such that subsequent use, during interactive runtime, has no or reduced positional errors and/or artifacts caused by positional errors.

A method for constructing a three-dimensional road network map is disclosed. The method includes: extracting a two-dimensional road based on a satellite remote sensing image; skeletonizing the two-dimensional road to obtain a two-dimensional road network map; and determining a height of each point in the two-dimensional road network map based on an elevation of three-dimensional point cloud data, to obtain the three-dimensional road network map.

A three-dimensional shape generation apparatus includes circuitry configured to perform one of a first processing or a second processing to generate three-dimensional shape information using point-group information indicating three-dimensional point groups as a generation processing based on an instruction operation received at an instruction reception screen, the instruction operation designating one of the first processing or the second processing. The circuitry performs all processes or a part of processes of the generation processing based on an operation input received at the instruction reception screen in response to the instruction operation designating the first processing. The circuitry performs all processes of the generation processing using a stored processing previously stored in a memory without using the operation input received at the instruction reception screen in response to the instruction operation designating the second processing.