Abstract: Embodiments of this application provide methods and apparatuses for compressing and decompressing a three-dimensional (3D) map, to reduce a data volume of a 3D map. The method for compressing a 3D map includes: obtaining a 3D map descriptor, where the 3D map descriptor corresponds to at least one 3D map point of the 3D map; performing binarization on a relationship between the 3D map descriptor and at least one preset threshold vector to obtain binary data; and performing quantization on a difference between the 3D map descriptor and the at least one preset threshold vector to obtain quantized data.

Abstract: A display device including a controller and a display panel. The controller receives original image data and output a display image signal. The display panel receives the display image signal and displays a display image corresponding to the display image signal. The controller includes an image shift controller and a memory. The image shift controller generates shifted image data by modulating the original image data to shift the display image sequentially along a preset shift path on the display panel. The memory stores a shift path value indicating a distance by which the display image has been shifted on the preset shift path. The image shift controller generates the display image signal by processing the shifted image data. When the display device is powered on, the image shift controller generates shifted image data corresponding to a shift path value stored in the memory.

Abstract: An information processing system includes: an acquisition unit configured to divide a target area into a plurality of meshes of a predetermined size, and acquire mesh data indicating the environment of the target area with height information about the plurality of meshes on a per-mesh basis; a first ray tracing unit configured to determine a two-dimensional ray trace from a transmission point to a reception point on the basis of the mesh data, a radio wave being transmitted at the transmission point, the radio wave being received at the reception point; a second ray tracing unit configured to determine a three-dimensional ray trace corresponding to the two-dimensional ray trace, on the basis of the mesh data, and height information about the transmission point and the reception point; and a radio field intensity calculation unit configured to calculate an intensity of the radio wave at the reception point, using one or more three-dimensional ray traces determined by the second ray tracing unit.

Abstract: The present disclosure provides a three-dimensional point cloud generation method, apparatus and an electronic device, and relates to the field of artificial intelligence technologies such as virtual reality, augmented reality and computer vision. A specific scheme includes: generating a two-dimensional (2D) depth map of a first three-dimensional (3D) image; calculating gradient values of multiple pixels in the 2D depth image, and performing filter process on the 2D depth image based on the gradient values of the multiple pixels to obtain a target depth image; and generating a 3D point cloud of the first 3D image based on the target depth map. According to the present disclosure, a generation efficiency of the three-dimensional point cloud can be improved.

Abstract: In implementations of techniques and systems for chamfer tools in volumetric sculpting software, a processing device receives an indication of a brushed region of a three-dimensional (3D) object to be chamfered and a chamfer size. A machine-learning model determines surface voxels in the brushed region representing an edge or corner voxel. The machine-learning model also determines parameters for the edge or corner voxel that define a chamfer plane and a chamfer coefficient. The chamfer plane is then moved in a direction normal to the chamfer plane by a distance based on the chamfer size and the chamfer coefficient. To generate a chamfer, the processing device performs an addition or subtraction operation on the 3D object to the moved chamfer plane for the edge or corner voxel.

Abstract: A method for creating a reference map representation using an artificial neural network, wherein the neural network comprises an encoder and a decoder. The method includes: receiving point cloud data representing a point cloud of the environment of the motor vehicle that was created on the basis of a detection of an environment of a motor vehicle by an environmental sensor of the motor vehicle; receiving state data representing a state of the environmental sensor during the detection of the environment by the environmental sensor; training the encoder and the decoder on the basis of the point cloud data and the state data.

Abstract: Techniques are described for automating the design and manufacture a dental restoration appliance for restoring the dental anatomy of a patient. For example, a system processes a digital three-dimensional (3D) model of a future (desired) dental anatomy of a patient, the future dental anatomy representing an intended shape of at least one tooth of the patient. The system includes a landmark identifier configured to automatically compute, based on the digital 3D model of the future dental anatomy of the patient, one or more landmarks of the future dental anatomy of the patient. The system also includes a custom feature generator configured to automatically generate, based on the one or more landmarks, one or more custom appliance features for a dental appliance for restoring the at least one tooth of the patient. The system further includes a memory device configured to store a digital model of the dental appliance.

Abstract: According to examples, a processor may identify locations on a digital 3D model of an object corresponding to positions at which features are to be added to the object. For each of a plurality of locations of the identified locations, the processor may identify displacements of each of a plurality of patch points around the location from the surface of the digital 3D model. The processor may also create a map of the digital 3D model that encodes the identified displacements of the patch points around the locations, in which a 3D fabrication system is to fabricate the object to include the plurality of features based on the digital 3D model and the created map.

Abstract: Methods and systems for generating geometric models of buildings from multiview imagery are provided. An example method involves displaying at least a first image and a second image that depict the building from different points of view, providing functionality for a user to annotate, with reference to at least the first image and the second image, a three-dimensional structure of a first building section of the building having a polygonal outline of a first substantially horizontal roof structure situated at a first height, and a three-dimensional structure of a second building section of the building having a polygonal outline of a second substantially horizontal roof structure situated at a second height lower than the first height. The polygonal outline of the second substantially horizontal roof structure is annotated by selecting a point located directly on one of the side walls of the first building section.

Abstract: An augmented reality (AR) device generates hand annotations for an image depicting a user's hand. The device includes a display, a processor, and a memory storing instructions for performing operations. The device performs a calibration operation to generate a 3-D model of the user's hand based on measurements. The calibration operation prompts the user to mimic the hand gesture presented by the animated virtual representation of a hand to obtain optimal images of the hand for hand pose estimation. A 3-D virtual representation of a hand in a hand pose corresponding with a hand gesture is generated based on the 3-D model. The device presents the 3-D virtual representation of the hand in AR via the display. During presentation, the device detects an input and captures an image of the user's hand positioned to correspond with the 3-D virtual representation. The captured image is stored with corresponding hand annotations based on the 3-D model.

Abstract: The present disclosure relates to a method for generating a design based on a learned condition performed by a design generation apparatus based on a learned condition. According to an embodiment of the present disclosure, the method may comprise acquiring an image from information including images and texts; learning features of the acquired image; extracting texts from the information and matching the extracted texts with the learned features of the image; learning a condition for a design to be generated based on the image through the matching; receiving texts inputted by a user for design image generation; identifying a condition corresponding to the user's texts; and generating a design image based on the identified condition.

Abstract: An information processing system includes: a first ray tracing unit configured to determine a two-dimensional ray trace from a transmission point at which a radio wave is transmitted, to a reception point at which the radio wave is received; a second ray tracing unit configured to determine a three-dimensional ray trace corresponding to the two-dimensional ray trace, using height information about the transmission point and the reception point; and a radio field intensity calculation unit that calculates the intensity of the radio wave at the reception point, using one or more three-dimensional ray traces determined by the second ray tracing unit.

Abstract: Systems and methods for generating a volumetric video in which all frames are temporally coherent.

Abstract: Determining an absolute depth estimate of a vehicle object in a 2D image captured by a camera includes receiving the 2D image captured by the camera. It further includes determining, at least in part by using a segmentation model, a plurality of pixels corresponding to the vehicle object in the 2D image. It furhter includes using a prediction model to determine, for each pixel in the plurality of pixels determined at least in part by using the segmentation model, a correspondng camera invariant distance value comprising a predicted height of a cross-section of the vehicle object that is visible within a given pixel. It further includes determining the absolute depth estimate of the vehicle object in the 2D image based on: a focal length of the camera that captured the 2D image; and an aggregation of the predicted heights determined for each of the pixels in the plurality of pixels.

Abstract: Systems and methods are disclosed for enhancing and developing a damage scene virtual reality (VR) visualization. Annotated immersive multimedia image(s) may be received from a first user, where the annotated immersive multimedia image(s) can be associated with a damage scene. A VR visualization of the annotated immersive multimedia image(s) may be rendered using a VR device associated with a second user. The VR visualization may be used to determine a damage amount, where the damage amount is determined from one or more damaged items identifiable in the annotated immersive multimedia image(s).

Abstract: Certain aspects and features of this disclosure relate to rendering images by training a neural material and applying the material map to a coarse geometry to provide high-fidelity asset encoding. For example, training can involve sampling for a set of lighting and camera configurations arranged to render an image of a target asset. A value for a loss function comparing the target asset with the neural material can be optimized to train the neural material to encode a high-fidelity model of the target asset. This technique restricts the application of the neural material to a specific predetermined geometry, resulting in a reproducible asset that can be used efficiently. Such an asset can be deployed, as examples, to mobile devices or to the web, where the computational budget is limited, and nevertheless produce highly detailed images.

Abstract: In an embodiment a method includes displaying, by an electronic device, a first sequence frame image of a prompt file in response to a first operation performed by a user and displaying, by the electronic device, a second sequence frame image of the prompt file in response to a second operation performed by the user, wherein the prompt file comprises at least first-part contour information and second-part contour information of a playing object, wherein the first sequence frame image is generated based on the first-part contour information of the playing object, and wherein the second sequence frame image is generated based on the second-part contour information of the playing object and comprises content of the first sequence frame image.

Abstract: Implementations are directed to generating corresponding three-dimensional (“3D”) rowview representation(s) of row(s) of an agricultural field at various time instance(s) to enable a human operator of the agricultural field to virtually traverse through the row(s) at the various time instance(s). In some implementations, the corresponding 3D rowview representation(s) can be generated based on corresponding vision data captured at the various time instance(s). The corresponding 3D rowview representation(s) can be generated based on processing the corresponding vision data. Further, the corresponding 3D rowview representation(s) can be provided to a client device of the human operator of the agricultural field to enable the human operator to virtually traverse through the row(s) of the agricultural field at the various time instance(s).

Abstract: A scene modeling system accesses a set of input two-dimensional (2D) images of a three-dimensional (3D) environment, wherein the input 2D images captured from a plurality of camera orientations. The environment includes first content. The scene modeling system applies a scene generation model to the set of input 2D images to generate a 3D remix scene. Applying the scene generation model includes configuring the scene generation model using at least a 2D discriminator and a 3D discriminator. Applying the scene generation model includes transmitting, for display via a user interface, the 3D remix scene. The 3D remix scene includes second content that is different from the first content.

Abstract: Various implementations of the present application set forth a method comprising generating, three-dimensional data and two-dimensional data representing a physical space that includes a real-world asset, generating an adaptable three-dimensional (3D) representation of the physical space based on the two-dimensional and three-dimensional data, where the adaptable 3D representation includes a plurality of coordinates representing different positions in 3D coordinate space corresponding to the physical space, transforming the adaptable 3D representation into geometry data comprising a set of vertices, faces comprising edges between pairs of vertices, and texture data, transmitting the geometry data to a remote device, wherein the remote device, constructs, based on the geometry data, the adaptable 3D representation of the physical space for display at a location of the remote device in a remote environment, and modifies, based on an input, at least one of a dimension or a position of the adaptable 3D representa