Abstract: One embodiment of a method for rendering one or more graphics images includes tracing one or more rays through a graphics scene; storing one or more hit points based on one or more coordinate frames associated with one or more voxels of a grid, where the one or more rays intersect one or more surfaces of geometry in the one or more voxels; and rendering one or more graphics images based on the one or more hit points that are stored.

Abstract: Systems and methods include a computer-implemented method for determining and storing a length scale dimension in spatial coordinate systems. A mapping of three-dimensional (3D) grid locations (x,y,z) relative to a continuous surface of the Earth is determined. The 3D grid locations have a grid point spacing in an (x,y) space defining map view coordinates of the continuous surface independent of an elevation z. A length scale (l) is determined for each 3D location. The length scale defines a 3D distance between 3D grid locations, and also defines a structural or geometrical length scale at the 3D location relevant to a given task. Then, (x,y,z,l) information is stored for each 3D grid location. The (x,y,z,l) information defines, for each (x,y) coordinate, a z-coordinate defining an elevation of the continuous surface at the (x,y) coordinate and the local length scale at the (x,y,z) coordinate.

Abstract: A computing device receives a request requesting a screening design for an experiment. The device obtains factors for screening. A first factor has a first set of candidate inputs comprising at least two candidate inputs for allowable inputs in the experiment. A second factor has a second set of candidate inputs comprising at least three candidate inputs for allowable inputs in the experiment. The device receives an indication of a run size. The device generates the screening design by assigning, based on the run size, the first factor to a two-level group that only allows at most two allowable inputs for the first factor in the experiment. The device generates the screening design by assigning, based on the initial run size, the second factor to a three-level group that only allows at most three allowable inputs for the second factor in the experiment. The device outputs the screening design.

Abstract: A control circuit accesses three-dimensional image information for a given three-dimensional object. The control circuit accesses a selection corresponding to a feature of the three-dimensional object, and then automatically generates a plurality of synthetic images of the three-dimensional object as a function of the three-dimensional and the selection of the aforementioned feature. By one approach, these synthetic images include supplemental visual emphasis corresponding to the aforementioned feature. The generated plurality of synthetic images can then be used as a training corpus when training a machine learning model.

Abstract: A computing device receives a request requesting a screening design for an experiment. The device obtains, for the screening design, factors for screening in the experiment. A first factor is a categorical factor with a first set of candidate inputs comprising at least three candidate inputs in the experiment. A second factor is a continuous factor with a second set of candidate inputs comprising candidate input options within range(s) for the second factor. The device receives an indication of a run size that is a user-requested quantity of runs for the screening design. The device generates the screening design based on the run size. The device outputs the screening design. The screening design comprises runs of a quantity based on the user-requested quantity of runs for the screening design. The runs comprise the first set of candidate inputs and at least three inputs of the second set of candidate inputs.

Abstract: A method includes accessing a first dataset including aerial imagery data and infrared data, accessing a second dataset including property boundary data, and identifying property boundaries associated with a geographic area. A plurality of models is applied to the datasets to identify and compute information of interest. Based on the first dataset and constrained by the property boundaries, a building detection model can be applied to identify a building, and a vegetation detection model can be applied to identify one or more areas of vegetation. An estimated distance can be determined between each of the areas of vegetation and the building as separation data, which can be compared to a defensible space guideline to determine a defensible space adherence score. A wildfire risk map can be generated, including the defensible space adherence score associated with the geographic area.

Abstract: A system includes: a feature module configured to generate a feature map based on a single image taken from a point of view (POV) including a human based on features of the human visible in the image and non-visible features of the human; a pixel features module configured to generate pixel features based on the feature map and a target POV; a feature mesh module configured to generate a feature mesh for the human based on the feature map; a geometry module configured to: generate voxel features based on the feature mesh; and generate a density value based on the voxel and pixel features; a texture module configured to generate RGB colors for pixels based on the density value and the pixel features; and a rendering module configured to generate a three dimensional rendering of the human from the target POV based on the RGB colors and the density value.

Abstract: A hardware-based traversal coprocessor provides acceleration of tree traversal operations searching for intersections between primitives represented in a tree data structure and a ray. The primitives may include opaque and alpha triangles used in generating a virtual scene. The hardware-based traversal coprocessor is configured to determine primitives intersected by the ray, and return intersection information to a streaming multiprocessor for further processing. The hardware-based traversal coprocessor is configured to omit reporting of one or more primitives the ray is determined to intersect. The omitted primitives include primitives which are provably capable of being omitted without a functional impact on visualizing the virtual scene.

Abstract: A generation apparatus includes an acquisition unit configured to acquire shape data representing a shape of a subject based on a plurality of captured images obtained by a plurality of imaging apparatuses capturing the subject in an imaging region, a first determination unit configured to determine a subject position which is a position of the subject in the imaging region, a second determination unit configured to determine a reference position serving as a reference of the position of the subject, and a generation unit configured to generate, based on the shape data acquired by the acquisition unit, a virtual viewpoint image in accordance with a deviation between the subject position determined by the first determination unit and the reference position determined by the second determination unit.

Abstract: A system and method for training a neural network. In some embodiments, the system includes: a graphics processing unit cluster; and a computational storage cluster connected to the graphics processing unit cluster by a cache-coherent system interconnect. The graphics processing unit cluster may include one or more graphics processing units. The computational storage cluster may include one or more computational storage devices. A first computational storage device of the one or more computational storage devices may be configured to (i) store an embedding table, (ii) receive an index vector including a first index and a second index; and (iii) calculate an embedded vector based on: a first row of the embedding table, corresponding to the first index, and a second row of the embedding table, corresponding to the second index.

Abstract: Aspects of the disclosure include methods, apparatuses, and non-transitory computer-readable storage mediums for point cloud compression. An apparatus includes processing circuitry that encodes information associated with a current point of a plurality of points of a point cloud. The plurality of points is partitioned into multiple bounding boxes. The processing circuitry determines whether a first size of a hash table is greater than or equal to a predetermined maximum size of the hash table. The processing circuitry removes information associated with non-boundary points in the multiple bounding boxes from the hash table based on the first size of the hash table being greater than or equal to the predetermined maximum size of the hash table. The processing circuitry stores the encoded information associated with the current point into the hash table.

Abstract: Aspects of the present disclosure involve a system for performing real-time in-painting using machine learning techniques. The system receives a video that includes a depiction of a real-world object in a real-world environment. The system accesses a segmentation associated with the real-world object and removes a depiction of the real-world object from a region of a first frame of the video. The system processes, by a machine learning model, the first frame and one or more previous frames of the video that precede the first frame to generate a new frame in which portions of the first frame have been blended into the region from which the depiction of the real-world object has been removed.

Abstract: A system includes at least one imaging sensor and a processor. The processor is configured to acquire detected data describing an environment of a vehicle using the at least one imaging sensor; derive reference data which describes the environment from a predefined map; compute difference data representing a difference between the detected data and the reference data; and transfer the difference data, wherein an image computed based on the difference data and the reference data represents the detected data.

Abstract: Apparatuses, systems, and techniques to perform a K-nearest-neighbor query. In at least one embodiment, a set of bounding boxes corresponding to a set of primitives is generated that allows the query to be solved using light transport simulation acceleration features of a GPU.

Abstract: An advertisement display system 1 acquires coordinates of the avatar and a size of the avatar in the virtual reality space and calculates a movement speed of the avatar by dividing a coordinate change amount per unit time of the avatar in the virtual reality space by a reference amount in accordance with the size of the avatar. In addition, the system 1 determines whether a point of view of the avatar in the virtual reality space is set to a first-person point of view or a third-person point of view. Furthermore, the system 1 performs control to differentiate, when an avatar is moving in the virtual reality space, a position where advertising is to be displayed in the virtual reality space with respect to the user operating the avatar in accordance with the movement speed of the avatar and the point of view of the avatar.

Abstract: Techniques applicable to a ray tracing hardware accelerator for traversing a hierarchical acceleration structure with reduced false positive ray intersections are disclosed. The reduction of false positives may be based upon one or more of selectively performing a secondary higher precision intersection test for a bounding volume, identifying and culling bounding volumes that degenerate to a point, and parametrically clipping rays that exceed certain configured distance thresholds.

Abstract: A generation device includes processing circuitry configured to receive a plurality of inputs of road map data including longitude/latitude data on lane information indicating a center line of a lane, longitude/latitude data on a road shoulder line, and longitude/latitude data on a lane marker, and refer to the road map data and generate a first polygon indicating a region of a lane based on a plurality of intersections on the lane marker or the road shoulder line crossed by a vertical line from the lane information.

Abstract: The present disclosure relates to a method and capturing arrangement for creating a three-dimensional model of a scene. The model comprises a three-dimensional space comprising a plurality of discrete three-dimensional volume elements (V1,1, V1,2) associated with three initial direction-independent color values and an initial opacity value. The method comprises obtaining a plurality of images of said scene and defining a minimization problem. Wherein the minimization problem comprises three residuals, one for each color value, wherein each residual is based on the difference between (a) the color value of each image element of each image and (b) an accumulated direction-independent color value of the volume along each ray path of each image element. The method further comprises creating the three-dimensional model of the scene by solving said minimization problem.

Abstract: Embodiments of the present disclosure provide an adjustment method and apparatus for a rendering hierarchy order, and the method includes: displaying, in response to an instruction triggered for a target moment position of a video editing timeline track, a rendering hierarchy adjustment panel corresponding to the target moment position, displaying identifiers of a plurality of editing material segments, whose time intervals covered on the video editing timeline track includes the target moment position, in the rendering hierarchy adjustment panel according to an initial order; moving for the identifier of the editing material segment and adjusting the display position of the identifier of the editing material segment, so that the identifiers of the plurality of editing material segments are displayed according to a target order; and then determining a target rendering hierarchy order of the plurality of editing material segments according to the target order.

Abstract: A mixed reality system that includes a device and a base station that communicate via a wireless connection The device may include sensors that collect information about the user's environment and about the user. The information collected by the sensors may be transmitted to the base station via the wireless connection. The base station renders frames or slices based at least in part on the sensor information received from the device, encodes the frames or slices, and transmits the compressed frames or slices to the device for decoding and display. The base station may provide more computing power than conventional stand-alone systems, and the wireless connection does not tether the device to the base station as in conventional tethered systems. The system may implement methods and apparatus to maintain a target frame rate through the wireless link and to minimize latency in frame rendering, transmittal, and display.