Abstract: A processor implemented method includes: calculating edge direction information of edge component pixels extracted from an input image; determining an inlier pixel among the edge component pixels based on the edge direction information and virtual horizontal line information; estimating a rotation reference point based on the inlier pixel; and correcting localization information of the apparatus based on the estimated rotation reference point.

Abstract: A method of embedding a vector representing an arrangement state of objects in a 3D space according to an embodiment includes dividing a target space into M×N subspaces extending from a first surface among planes forming the target space having a cuboid shape, calculating, for each of the subspaces, a feature representing an object arrangement state in the subspace, inputting data representing the feature of each of the subspaces into an artificial neural network and obtaining, from the artificial neural network, an embedding vector representing an arrangement state of an object in the target space.

Abstract: The present invention discloses a positive azimuth towing guidance method for road rescue equipment based on license plate corner features. The method combines a structure of the road rescue equipment and characteristics of a positive azimuth towing operation. First, an image of an operation area is collected by installing a camera, and grayscale processing and Gaussian smooth filtering are performed on the image; corner detection is performed on the smoothed grayscale image, and preference is implemented according to corner strengths; hierarchical clustering is performed on the preferred corners; an effective corner set of license plate characters is sorted out to implement license plate locating; and then towing guidance is implemented according to a license plate locating result, to improve the rescue efficiency of the road rescue equipment.

Abstract: Techniques and architecture are disclosed for mobile transport systems configured to determine vehicle positions within an area using light-based communication signals. The system includes a plurality of luminaires located in an area and configured to transmit luminaire position data recognizable by a sensor disposed on a vehicle. The sensor receives an image of a luminaire including a light-based communication signal encoded with luminaire position data. Luminaire position data can be combined with luminaire layout information to determine a known location of the luminaire. A vehicle position relative to the known luminaire location can be determined based on mathematical relationships. Vehicle orientation relative to the area can be determined based an asymmetric fiducial pattern or multiple known luminaire locations. The system can combine a vehicle position relative to a known luminaire location with vehicle orientation relative to the area to determine a vehicle position relative to the area.

Abstract: An image processing method is provided for an electronic device. The method includes obtaining a target image comprising a target object; recognizing target two-dimensional location coordinates of the target object in the target image and a target attribute type corresponding to the target object; and obtaining a target three-dimensional point cloud associated with the target image. The method also include, according to a mapping relationship between the target three-dimensional point cloud and all pixels in the target image, obtaining three-dimensional location coordinates corresponding to pixels in the target two-dimensional location coordinates, as target three-dimensional location coordinates; determining a setting region in three-dimensional map data according to the target three-dimensional location coordinates; and setting the target attribute type for the target object in the setting region. Electronic device and non-transitory computer-readable storage medium counterparts are also contemplated.

Abstract: To detect a discrimination error in a type of an object. A calculation system includes a first device and a second device. The first device includes: a first object map generation unit configured to calculate, using first image information that is image information acquired by the first device, a first object map indicating a type of an object and a position of the object; and a first communication unit configured to transmit the first object map to the second device. The second device includes: a second object map generation unit configured to calculate, using second image information that is image information acquired by the second device, a second object map indicating a type of an object and a position of the object; and a comparison unit configured to compare the first object map and the second object map.

Abstract: A video analysis apparatus is connected to an image capturing apparatus including a plurality of cameras, in which the video analysis apparatus, for a plurality of persons in images captured by the cameras, tracks at least one of the plurality of persons, detects a plurality of parts of the tracked person, and on the basis of information defining scores used to determine a best shot of each of the parts, computes a score for the part for each of frames of videos from security cameras, compares, for each of the parts, the scores for each part computed for each of the frames to determine a best shot of each of the parts, stores, for each of the parts, a feature value in association with the part, and compares the feature values of each of the parts of the plurality of persons to determine identity of the persons.

Abstract: A method and system for calibrating a lens. The method includes defining a plurality of omni-symmetrical regions within the lens, determining one or more localized lens parameters associated with each of the plurality of omni-symmetrical regions, and defining a localized set of calibration parameters for each of the plurality of omni-symmetrical region. The localized set of calibration parameters may then be employed in a computational image application.

Abstract: An image synthesizing method includes capturing a first image using a reference small aperture size; capturing a second image using a reference large aperture size; obtaining one or more reference color weights according to a corresponding pixel of the first image and adjacent pixels of the corresponding pixel of the first image and a corresponding pixel of the second image; obtaining an associated distance by looking up an association table according to the one or more reference color weights; obtaining one or more associated color weights by looking up the association table according to the associated distance and an expected aperture; and obtaining a color value of a corresponding pixel of a synthesized image, by applying weighting to the corresponding pixel of the first image and the adjacent pixels of the corresponding pixel of the first image with the one or more associated color weights.

Abstract: A deep learning method employs a neural network having three sub-nets to classify and retrieve the most similar 3D model of an object, given a rough 3D model or scanned images. The most similar 3D model is present in a database and can be retrieved to use directly or as a reference to redesign the 3D model. The three sub-nets of the neural network include one dealing with object images and the other two dealing with voxel representations. Majority vote is used instead of view pooling to classify the object. A feature map and a list of top N most similar well-designed 3D models are also provided.

Abstract: Systems, apparatuses and methods may provide for technology that partitions a high dynamic range (HDR) image into a plurality of regions and determines, on a per region basis, a luminance level of the HDR image. Additionally, the technology may select, on the per image basis, a encoding amount for each region in the plurality of regions based on the luminance level.

Abstract: A system comprises an encoder configured to compress attribute information and/or spatial information for a point cloud and/or a decoder configured to decompress compressed attribute and/or spatial information for the point cloud. The encoder is configured to convert a point cloud into an image based representation. The encoder packs patch images into an image frame and fills empty spaces in the image frame with a padding. Various compression strategies may be used to encode an occupancy map and related block-to-patch information indicating which portions of the image frame correspond to which packed patches. Packed image frames comprising patches and padding, along with an encoded occupancy map and related block-to-patch information are sent to a decoder. The decoder is configured to generate a decompressed point cloud based on the packed image frames, compressed occupancy map, and related block-to-patch information.

Abstract: When a normal inspection and an inspection through deep learning processing is applicable, high inspection accuracy is obtained while reducing a processing time. The normal inspection processing is applied to a newly acquired inspection target image, the non-defective product determination or the defective product determination is confirmed for the inspection target image having the characteristic amount with which the non-defective product determination or the defective product determination is executable based on the characteristic amount within the inspection target image and the threshold for confirming the non-defective product determination or the threshold for confirming the defective product determination.

Abstract: A method for aligning a real-world object with a virtual object includes capturing images, video, or both of the real-world object from a first viewpoint and from a second viewpoint. The first and second viewpoints are different. The method also includes simultaneously superimposing the virtual object at least partially over the real-world object from the first viewpoint in a first augmented reality (AR) display and from the second viewpoint in a second AR display based at least in part on the images, video, or both. The method also includes adjusting a position of the real-world object to at least partially align the real-world object with the virtual object from the first viewpoint in the first AR display and from the second viewpoint in the second AR display.

Abstract: A method of imaging includes obtaining projection data for an object representing an intensity of radiation detected along a plurality of rays through the object, obtaining an outline of the object via a secondary imaging system, the secondary imaging system using non-ionizing radiation, determining, based on the outline, a model and model parameters for the object, calculating, based on the model and the model parameters, a volumetric attenuation map for the object, and reconstructing, based on the projection data and the volumetric attenuation map, an attenuation-corrected volumetric image.

Abstract: A method for evaluating image quality is provided. The method may include: obtaining an image, the image including a plurality of elements, each element of the plurality of elements being a pixel or voxel, each element having a gray level; determining, based on a maximum gray level of the plurality of elements, one or more thresholds for segmenting the image; determining one or more sub-images of a region of interest by segmenting, based on the one or more thresholds, the image; and determining, based on the one or more sub-images of the region of interest, a quality index for the image.

Abstract: An image processing apparatus includes a combining unit configured to combine a plurality of images to generate a combined image, and a control unit configured to record the combined image generated by the combining unit on a recording medium, and configured to: if the combined image, generated by the combining unit, is generated by combining a plurality of captured images, and: if the combined image satisfies a predetermined condition, attach first type data to the recorded combined image, and if the combined image does not satisfy the predetermined condition, attach second type data to the recorded combined image. If the combined image, generated by the combining unit, is generated by combining a plurality of captured images recorded on the recording medium, the second type data is attached regardless of whether the predetermined condition is satisfied.

Abstract: Systems and methods for displaying a virtual reticle in an augmented or virtual reality environment by a wearable device are described. The environment can include real or virtual objects that may be interacted with by the user through a variety of poses, such as, e.g., head pose, eye pose or gaze, or body pose. The user may select objects by pointing the virtual reticle toward a target object by changing pose or gaze. The wearable device can recognize that an orientation of a user's head or eyes is outside of a range of acceptable or comfortable head or eye poses and accelerate the movement of the reticle away from a default position and toward a position in the direction of the user's head or eye movement, which can reduce the amount of movement by the user to align the reticle and target.

Abstract: Systems and methods provide snap-line generation operations for live multi-path glyphs. In this regard, an attribute extraction component accesses a glyph description of a live multi-path glyph to determine and store several attributes related to at least one sub-path of the live multi-path glyph. The stored attributes are accessed by a snap-line generation component and used to determine whether an external object is moved near the live multi-path glyph. When moved near one another, a snap-line is generated along sub-path planes of the external object and the live multi-path glyph. Generated snap-lines are used to aid a user in aligning the two objects.

Abstract: The disclosure relates to a method and a device for generating a sticker. The method and device can receive a sticker configuration file from a server, where the sticker configuration file includes configuration parameters, and the configuration parameters represent display styles of a background image element and at least one text element in a target sticker to be generated; acquire a sticker background image and sticker characters input by a user as the background image element and the at least one text element respectively; and generate the target sticker according to the background image element, at least one the text element and the configuration parameters.

Abstract: A system and method of creating customized characters and selectively displaying them in an electronic display, such as an augmented reality or virtual reality display is provided. A digital character may be provided by a character provider for customization by others using the system. Such customizations may be instantiated in user devices that provide electronic displays. Instantiation of the custom digital character may be conditioned on one or more trigger conditions, which may be specified by the character customizer. For example, a digital character customized using the system may be conditioned on triggering events in the real-world or in a virtual world. When a relevant triggering condition is satisfied at a user device, the custom character (i.e., information for instantiating the custom character) may be transmitted to that user device. In this manner, the system may push custom characters to user devices that satisfy the triggering condition.

Abstract: A server machine modifies an augmented reality (AR) object in response to fulfillment of a condition. The machine provides, to a user device, object data that defines the AR object. The object data specifies a physical geolocation of the AR object, a presentation attribute of the AR object, a conditional modification program, and a trigger condition for execution of the conditional modification program. The object data causes the user device to present the AR object with a first appearance, located at the physical geolocation. The machine detects fulfillment of the trigger condition, and in response, the machine executes the conditional modification program. This modifies the object data by modifying the presentation attribute. The machine provides, to the user device, the modified object data, which causes the user device to present the AR object with a second appearance based on the modified presentation attribute.

Abstract: In accordance with an aspect of the present disclosure, there is provided a method for simplifying three-dimensional mesh data using three-dimensional mesh data simplification device. The method comprises, determining a vertex or an edge of three-dimensional mesh data to be deleted based on animation information including skin weight values and geometric information of the three-dimensional mesh data; and simplifying the three-dimensional mesh data by deleting the vertex or the edge.

Abstract: A method for generating a three-dimensional local human body model are provided in embodiments of the disclosure. The method for generating a three-dimensional local human body model includes: acquiring bone parameter adjustment information of a three-dimensional local human body model; adjusting bone parameters of the three-dimensional local human body model according to the bone parameter adjustment information; and generating the three-dimensional local human body model according to the adjusted bone parameters.

Abstract: A system and method for controlling a virtual light source includes generating a volume-rendered image including a rendered object, controlling a shading and an illumination of the volume-rendered image based on a position of the virtual light source with respect to the rendered object, and receiving a control input from a user interface. The system and method includes automatically moving the virtual light source to an updated position along a height contour established with respect to a surface of a rendered object in the volume-rendered image in response to the control input. The system and method includes calculating an updated shading and an updated illumination based on the updated virtual light source position.

Abstract: An image processing device includes a three-dimensional graphics processor that generates three-dimensional drawing data based on a three-dimensional drawing command, a command converter that converts the three-dimensional drawing command into a two-dimensional drawing command, a two-dimensional graphics processor that generates two-dimensional drawing data based on the converted two-dimensional drawing command and a display controller that displays the two-dimensional drawing data in place of the three-dimensional drawing data when a load of the three-dimensional graphics processor is high.

Abstract: The present invention relates generally to the field of computer vision and computer graphics to produce full body renderings of a person for varying person pose and camera positions and, in particular, to a system and method for synthesizing 2-D image of a person.

Abstract: In various examples, the actual spatial properties of a virtual environment are used to produce, for a pixel, an anisotropic filter kernel for a filter having dimensions and weights that accurately reflect the spatial characteristics of the virtual environment. Geometry of the virtual environment may be computed based at least in part on a projection of a light source onto a surface through an occluder, in order to determine a footprint that reflects a contribution of the light source to lighting conditions of the pixel associated with a point on the surface. The footprint may define a size, orientation, and/or shape of the anisotropic filter kernel and corresponding filter weights. The anisotropic filter kernel may be applied to the pixel to produce a graphically-rendered image of the virtual environment.

Abstract: Raytracing can be used to generate high quality, physics-based water caustics patterns in real time. A caustics map is generate to represent locations and normals of points across a water surface. Rays from a light source that are reflected and refracted from these points, as determined by the locations and normals, and can generate hit points on a surface. Neighboring points can be used to help determine the resulting caustics pattern. In one embodiment, information for neighboring points in the caustics map can be used to generate scale factors for geometric regions to be projected onto the surface for each hit point. In another embodiment, these points serve as vertices of a caustic mesh that can be projected onto the surface, where the brightness at a primitive is determined by the size of the primitive area defined by the vertices of the caustics mesh.

Abstract: Disclosed techniques relate to ray intersection processing for ray tracing. In some embodiments, ray intersection circuitry traverses a spatially organized acceleration data structure and includes bounding region circuitry configured to test, in parallel, whether a ray intersects multiple different bounding regions indicated by a node of the data structure. Shader circuitry may execute a ray intersect instruction to invoke traversal by the ray intersect circuitry and the traversal may generate intersection results. The shader circuitry may shade intersected primitives based on the intersection results. Disclosed techniques that share processing between intersection circuitry and shader processors may improve performance, reduce power consumption, or both, relative to traditional techniques.

Abstract: An apparatus and method for performing BVH compression and decompression concurrently with stores and loads, respectively.

Abstract: The disclosure presents a technique for utilizing ray tracing to produce high quality visual scenes with shadows while minimizing computing costs. The disclosed technique can lower the number of rays needed for shadow region rendering and still maintain a targeted visual quality for the scene. In one example, a method for denoising a ray traced scene is disclosed that includes: (1) applying a pixel mask to a data structure of data from the scene, wherein the applying uses the scene at full resolution and pixels at the edge of a depth boundary change are identified using the pixel mask, (2) generating a penumbra mask using the data structure, (3) adjusting HitT values in the packed data buffer utilizing the penumbra mask, and (4) denoising the scene by reducing scene noise in the data of the data structure with adjusted HitT values.

Abstract: In an example a method includes identifying, by a processor, in a data model of at least a portion of a three-dimensional object, an object property associated with a location in the three-dimensional object. A data model of a virtual build volume comprising at least a portion of the three-dimensional object may be generated in which an association with an object property is dispersed beyond the location.

Abstract: A volumetric data structure models a particular volume representing the particular volume at a plurality of levels of detail. A first entry in the volumetric data structure includes a first set of bits representing voxels at a first level of detail, the first level of detail includes the lowest level of detail in the volumetric data structure, values of the first set of bits indicate whether a corresponding one of the voxels is at least partially occupied by respective geometry, where the volumetric data structure further includes a number of second entries representing voxels at a second level of detail higher than the first level of detail, the voxels at the second level of detail represent subvolumes of volumes represented by voxels at the first level of detail, and the number of second entries corresponds to a number of bits in the first set of bits with values indicating that a corresponding voxel volume is occupied.

Abstract: Encoding/decoding data representative of a 3D representation of a scene according to a range of points of view can involve generating a depth map associated with a part of the 3D representation according to a parameter representative of a two-dimensional parameterization associated with the part and data associated with a point included in the part, wherein the two-dimensional parameterization can be responsive to geometric information associated with the point and to pose information associated with the range of points of view. A texture map associated with the part can be generated according to the parameter and data associated with the point. First information representative of point density of points in a part of the part can be obtained. The depth map, texture map, parameter, and first information can be included in respective syntax elements of a bitstream.

Abstract: The disclosed embodiments include systems and methods to evaluate formation resistivity. The method includes obtaining, from a downhole tool deployed in a borehole, a plurality of values of formation resistivity of a downhole formation proximate the borehole. The method also includes generating a plurality of two-dimensional renderings of the formation resistivity based on the plurality of values, wherein each two-dimensional rendering of the plurality of two-dimensional renderings illustrates an inversion of the formation resistivity along a plane of the downhole formation. The method further includes generating a volumetric rendering of the formation resistivity of the downhole formation from the plurality of two-dimensional renderings, wherein the volumetric rendering comprises a plurality of two-dimensional planes, and wherein each two-dimensional plane of the plurality of two-dimensional planes illustrates an inversion of formation resistivity along the respective plane.

Abstract: Some embodiments provide a non-transitory machine-readable medium that stores a program executable by at least one processing unit of a device. The program provides a display area for viewing a 3D model that includes a plurality of three-dimensional (3D) objects. The program further provides a tool for viewing 3D objects in the 3D model. The program also determines a 3D object in the plurality of 3D objects in the 3D model to hide. The program further hides the determined 3D object in the 3D model.

Abstract: The method for virtual interaction with a three-dimensional indoor room includes: generating a virtual room model, generating a virtual room visual representation, providing the room data to a display device, receiving a virtual object selection, rendering an updated virtual room visual representation based on the virtual object, and providing the updated virtual room visual representation to the display device. The method can optionally include updating virtual room S700. A system for virtual interaction with a three-dimensional indoor room includes: a backend platform and a front end application.

Abstract: The disclosed technology presents a device and a method, respectively, for generating an augmented reality (AR) image (or AR video). The device configures to obtain a first image of a scene, obtain a second image of an object, obtain a depth range, and capture a third image of only the parts of the scene inside the depth range. Further, the device configures to generate occlusion information based on the third image, and overlay the first image and the second image based on the occlusion information to generate the AR image.

Abstract: A system, method and computer program product are provided for generating line-of-sight information using imagery. In some aspects, the method includes receiving a raster image depicting at least one object in a region of interest, and measuring a shadow of the at least one object in the raster image. The method also includes determining an angle of the sun from the shadow of the at least one object, and creating a plurality of virtual shadows for the at least one object using the angle of the sun. The method further includes generating line-of-sight information for the at least one object based on an intersection of the plurality of virtual shadows with objects in the region of interest.

Abstract: Autonomous vehicles and techniques that can be utilized to compress point cloud data and operate on compressed point cloud data are provided. An autonomous vehicle can include a data compression system can configure point cloud data according to a collection of three-dimensional (3D) tiles representative of the region. Each 3D tile can include a portion of the cloud point data, where each point vector in the portion of the cloud point data can be configured relative to a position vector of the 3D tile defined in a coordinate system of the collection of 3D tiles. The data compression system can utilize a fixed-point Q-format representation based on a defined number of bits to compress at least a portion of the point cloud data. The autonomous vehicle also can include a control system that can operate mathematically on compressed point cloud data, without reliance on prior decompression.

Abstract: A method, computer-readable storage medium, and device for generating a character model. The method comprises: receiving an input image of a reference subject; processing the input image to generate a normalized image; identifying a set of features present in the normalized image, wherein each feature in the set of features corresponds to a portion of a head or body of the reference subject; for each feature in the set of features, processing at least a portion of the normalized image including the feature by a neural network model corresponding to the feature to generate a parameter vector corresponding to the feature; and combining the parameter vectors output by respective neural network models corresponding to respective features in the set of features to generate a parameterized character model corresponding to reference subject in the input image.

Abstract: An example of a computer-readable medium is provided to store machine-readable instructions. The instructions may cause a processor to receive three-dimensional (3D) object representations of an object. The distance between points of a grid applied to 3D object representation and random points on the 3D object representation may provide a sampling error used in calculating a modeling accuracy between the 3D object representations.

Abstract: An object management system, an aircraft design system, and a method for managing an object. A three-dimensional environment with a model of an object and an avatar representing a human operator from a viewpoint relative to the avatar is displayed on a display system. A motion of the human operator is detected. An interaction between the avatar and the model of the object is identified in real time using information about motions of the human operator that are detected in real time. The interaction changes a group of dimensions in the model of the object. Further, the interaction between the avatar and the model of the object in the three-dimensional environment is displayed on the display system, enabling design changes in the model of the object made by the human operator.

Abstract: There is provided an information processing apparatus to bring a three-dimensional model generated in accordance with observation information, closer to a real object. The information processing apparatus includes: a control section configured to allocate, to a second three-dimensional model being at least a partial three-dimensional model included in a first three-dimensional model, a definite shaped model having a predetermined shape corresponding to a shape of the second three-dimensional model.

Abstract: A display device and a method thereof are provided. The display device includes a display, a communicator, and a processor to display a virtual reality (VR) content on the display, in response to a trigger signal and motion information to change a display viewpoint area of the VR content being received from a user terminal device through the communicator, control the display to display by changing a display viewpoint area of the VR content based on first motion direction included in the motion information until a predetermined signal is received. In response to the predetermined signal being received from the user terminal device, control the display to terminate changing the display viewpoint area.

Abstract: A preferred method for dynamically displaying virtual and augmented reality scenes can include determining input parameters, calculating virtual photometric parameters, and rendering a VAR scene with a set of simulated photometric parameters.

Abstract: A rear-facing camera captures a live-action video image while a front-facing camera captures an image of a distributor. The avatar controller controls an avatar based on the image of the distributor captured by the front-facing camera, and move the position of the avatar in the real space coordinate system according to the position of the video synthesis device in the real space coordinate system. A synthesizer arranges the avatar in a predetermined position of a real space coordinate system and synthesizes the avatar with the live-action video image.

Abstract: One or more computing devices, systems, and/or methods for presenting augmented reality (AR) interfaces are provided. For example, a first object corresponding to a representation of content in an AR interface may be presented. Responsive to receiving a selection of the first object, a first graphical object corresponding to the content may be presented. An AR interface comprising a real time view of a camera of the device may be presented. A first graphical representation of the first graphical object may be presented overlaid on the real time view of the camera of the device. A second graphical representation of the first graphical object comprising the graphical modification and a second graphical object associated with information corresponding to the content may be presented overlaid on the real time view of the camera of the device. Responsive to receiving a selection of the second graphical object, the information may be presented.

Abstract: A system and method for implementing an acceleration structure are disclosed. The method includes: determining a position of an entity; mapping the position to a 3D (three-dimensional) voxel grid coordinate in the acceleration structure, which comprises a central grid structure and six additional grid structures each comprising a set of voxels; determining a first offset value corresponding to the 3D voxel grid coordinate that corresponds to either the central grid structure or one of the six additional grid structures; and determining a second offset value corresponding to the 3D voxel grid coordinate that corresponds to a particular voxel within either the central grid structure or one of the six additional grid structures corresponding to the first offset value, wherein data corresponding to the entity is stored in memory a location based on the first offset value and the second offset value.