Abstract: Apparatuses, systems, and techniques to train one or more neural networks using stratified sampled training data parameters. In at least one embodiment, one or more stochastic training data parameters may be stratified sampled from one or more sampling ranges to compute a gradient for updating the one or more neural networks.

Abstract: Apparatuses, systems, and techniques to render computer graphics. In at least one embodiment, a first one or more lights are selected from among lights in a virtual scene to be rendered as a frame of graphics, and a second one or more lights are selected from among lights used to render one or more pixels in at least one of a prior frame or the current frame. A pixel of the current frame is rendered using the first and second one or more lights, and a light is selected for reuse in rendering a subsequent frame from among the first and second one or more lights.

Abstract: Disclosed subject matter relates generally to predictive graphics processing for interactive content.

Abstract: A splat generation system and associated methods implement a modified splatting pipeline that quantifies loss at each splat generation iteration without rendering the splats generated at each iteration or comparing the resulting visualization against reference images of a 3D asset. The system receives the 3D asset and generates a splat representation of the 3D asset. The system compares geometric similarity between different sets of generated splats in different regions of the splat representation and different sets of 3D primitives in corresponding regions of the 3D asset. The system adjusts a set of splats in response to the geometric similarity between the set of splats and a set of 3D primitives in corresponding regions not satisfying a threshold, and reconstructs the 3D asset with an acceptable amount of loss based on the adjustments made to the splats on a region-by-region basis.

Abstract: Embodiments described herein provide a 3D generation system from a single RGB image of an object by inferring the hidden 3D structure of objects based on 2D priors learnt by a generative model. Specifically, the 3D generation system may reconstruct the 3D structure of an object from an input of a single RGB image and optionally an associated depth estimate. For example, a radiance field is formulated to depict the input image in one viewpoint of the target 3D object, based on which other viewpoints of the 3D object can be inferred. Based on the visible surface depicted by the input image, points between the reference camera and the surface are assigned with zero density, and points on the surface are assigned with high density and color equal to the corresponding pixel in the input image.

Abstract: Embodiments described herein provide a 3D generation system from a single RGB image of an object by inferring the hidden 3D structure of objects based on 2D priors learnt by a generative model. Specifically, the 3D generation system may reconstruct the 3D structure of an object from an input of a single RGB image and optionally an associated depth estimate. For example, a radiance field is formulated to depict the input image in one viewpoint of the target 3D object, based on which other viewpoints of the 3D object can be inferred. Based on the visible surface depicted by the input image, points between the reference camera and the surface are assigned with zero density, and points on the surface are assigned with high density and color equal to the corresponding pixel in the input image.

Abstract: A method for assessing the physically based simulation quality of a glazed objects, in particular glazing buildings or glazing vehicles. A hyperspectral canopy hemispherical image of a real sky at given daylight conditions and an image of glazed object, both synchronously acquired, are processed into a physically based rendering to evaluate the colour difference between glazing parts of a rendered representative glazed object and the glazed object of the provided image. An outstanding advantage is a realistic rendering of a glazed object by an accurate rendering of reflecting effects coming from the sky onto glazed parts whatever the viewing angles.

Abstract: Techniques for improving how LSR is performed are disclosed. A service accesses a depth image, a pose correction matrix, and a color image. The service extracts a carrier geometry from the depth image. The service forward projects the LSR carrier geometry by multiplying each vertex of the LSR carrier geometry with the pose correction matrix. While the GPU is operating in a shadow map mode, the service causes the GPU to perform a rasterization process to produce a UV map and a Z buffer. The service discards the UV map, resulting in the per pixel UV corrections included in the UV map also being discarded. The service recovers the per pixel UV corrections using the Z buffer. The service uses the recovered per pixel UV corrections to resample the color image, resulting in generation of a corrected color image.

Abstract: The present disclosure describes techniques for generating three-dimensional models using machine learning models. A two-dimensional (2D) image is input into a machine learning model. The machine learning model is configured to generate three-dimensional (3D) models with accurate geometry and detailed textures. A set of multi-view images is generated based at least in part on the 2D image by a first sub-model of the machine learning model. The first sub-model comprises a multi-level image prompt controller configured to implement hierarchical controls over generating multi-view images by the first sub-model based at least in part on an input image. A 3D model is generated based at least in part on the set of multi-view images by a second sub-model of the machine learning model. The second sub-model is configured to implement a background alignment and a camera alignment for improving quality and geometric accuracy of the generated 3D models.

Abstract: This disclosure teaches techniques, devices, and systems for automatically modifying a 3-Dimensional (3D) object model. The apparatus comprising a memory and a processing device coupled to the memory. The processor and the memory are configured to obtain an input 3D object model comprising an input 3D mesh, analyze the input 3D mesh to identify unprintable overhangs that have an angle over an overhang angle limit and identify vertices associated with the unprintable overhangs as active vertices, and perform an optimization process to move the active vertices to eliminate the unprintable overhangs and generate a modified 3D mesh. The optimization process includes a gradient descent search algorithm performed on an objective function comprising a cost function configured to minimize a difference between the input 3D mesh and the modified 3D mesh subject to an overhang constraint related to the overhang angle limit.

Abstract: A system and method for coherency gathering for rays in a ray tracing system. The ray tracing system uses a hierarchical acceleration structure comprising a plurality of nodes including upper level nodes and lower level nodes. For each instance where one of the lower level nodes is a child of one of the upper level nodes, an instance transform is defined, specifying the relationship between a first coordinate system of the upper level node and the second coordinate system for that instance of the lower level node. The system provides an instance transform cache for storing a plurality of these instance transforms while conducting intersection testing.

Abstract: An image processing apparatus comprises a receiver (201) for receiving an image signal which comprises at least an encoded image and a target display reference. The target display reference is indicative of a dynamic range of a target display for which the encoded image is encoded. A dynamic range processor (203) generates an output image by applying a dynamic range transform to the encoded image in response to the target display reference. An output (205) then outputs an output image signal comprising the output image, e.g. to a suitable display. The dynamic range transform may furthermore be performed in response to a display dynamic range indication received from a display. The invention may be used to generate an improved High Dynamic Range (HDR) image from e.g. a Low Dynamic Range (LDR) image, or vice versa.

Abstract: An electronic device may include a camera module, a processor operably connected to the camera module, and a memory operably connected to the processor, wherein the memory may store one or more instructions for allowing, during the execution thereof, the processor to identify a first interaction for a first application being executed, form a first application object for outputting an execution screen of the first application according to the identification of the first interaction, output an augmented reality space on the basis of an actual space recognized through the camera module, and output the first application object in the augmented reality space.

Abstract: An image processing apparatus comprises a receiver (201) for receiving an image signal which comprises at least an encoded image and a target display reference. The target display reference is indicative of a dynamic range of a target display for which the encoded image is encoded. A dynamic range processor (203) generates an output image by applying a dynamic range transform to the encoded image in response to the target display reference. An output (205) then outputs an output image signal comprising the output image, e.g. to a suitable display. The dynamic range transform may furthermore be performed in response to a display dynamic range indication received from a display. The invention may be used to generate an improved High Dynamic Range (HDR) image from e.g. a Low Dynamic Range (LDR) image, or vice versa.

Abstract: A human body pose estimation method based on radio frequency heatmap data enhancement is provided, including the following steps: firstly, obtaining the mesh data of human body pose, simulating a radar by using a physical optics method, and obtaining human body mesh features including a radar cross section by irradiating a human body mesh model; secondly, processing the human body mesh features including radar cross section to obtain preliminary simulated radar heatmaps; then inputting the preliminary simulated radar heatmaps into a heatmap conversion network map2map based on U-net, outputting synthetic radar heatmaps, and performing training; finally, combining the synthetic radar heatmaps with real radar heatmaps to construct a mixed data set, obtaining a human body pose skeleton through a human body pose estimation network based on the radar heatmaps, and performing training to complete the human body pose estimation.

Abstract: A method, device, and computer-readable storage medium for generating an occlusion value for a pixel. The method includes: selecting a pixel in an image of a scene; identifying at least one direction to cast a fixed-distance ray from a location on an object corresponding to the pixel based on an orientation of a surface of the object at the location; in response to determining that screen space occlusion data is available in the at least one direction, obtaining an occlusion value for the pixel from at least one lighting probe based on the at least one direction; and in response to determining that screen space occlusion data is not available in the at least one direction, obtaining the occlusion value for the pixel based on performing ray tracing from the location on the object corresponding to the pixel.

Abstract: A computer-implemented method for rendering medical volume data including first volume data is described. The method comprises: performing a first volume rendering process on the first volume data to generate a first volume rendering of the first volume data. The first volume rendering process includes determining first depths of respective locations of the first volume rendering and storing the depths in association with the respective locations. The method also comprises performing a further process, including a second volume rendering process, on the medical volume data, to generate a second volume rendering, using the determined first depths and respective locations.

Abstract: An information processing system comprises processing circuitry configured to generate first information for displaying a video on an information processing apparatus; transmit the first information to the information processing apparatus; receive operation information regarding a plurality of operations performed on a first icon to take an action on the video displayed on the information processing apparatus; and determine, based on the operation information, a display mode of an object corresponding to the first icon, wherein the processing circuitry generates the first information for displaying the object on the video in the display mode.

Abstract: Assets in a virtual environment are assigned to locations based on attributes of the locations, such as light level, temperature, etc. Attributes are determined based on simulating the virtual environment. The virtual environment can be simulated over a period of time, during which multiple attribute values can be recorded, resulting in an aggregated attribute value to better enable more immersive asset placement.

Abstract: Ray tracing units, processing modules and methods are described for generating one or more reduced acceleration structures to be used for intersection testing in a ray tracing system for processing a 3D scene. Nodes of the reduced acceleration structure(s) are determined, wherein a reduced acceleration structure represents a subset of the 3D scene. The reduced acceleration structure(s) are stored for use in intersection testing. Since the reduced acceleration structures represent a subset of the scene (rather than the whole scene) the memory usage for storing the acceleration structure is reduced, and the latency in the traversal of the acceleration structure is reduced.