Abstract: Apparatus and method for determining a quality score for virtual video cameras. For example, one embodiment comprises: a region of interest (ROI) detector to detect regions of interest within a first image generated from a first physical camera (PCAM) positioned at first coordinates; virtual camera circuitry and/or logic to generate a second image positioned at the first coordinates; image comparison circuitry and/or logic to establish pixel-to-pixel correspondence between the first image and the second image; an image quality evaluator to determine a quality value for the second image by evaluating the second image in view of the first image.

Abstract: A server includes a processor and a memory storing instructions that, when executed by the processor, cause the server to receive two-dimensional (2D) images, analyze the images using a trained deep network to generate points, process the labeled points to identify tie points, and combine the 2D dimensional images into a three-dimensional (3D) point cloud using structure-from-motion. A method for generating a semantically-segmented 3D point cloud from 2D data includes receiving 2D images, analyzing the images using a trained deep network to generate labeled points, processing the points to identify tie points, and combining the 2D images into a 3D point cloud using structure-from-motion. A non-transitory computer readable storage medium stores executable instructions that, when executed by a processor, cause a computer to receive 2D images, analyze the images using a trained deep network to generate labeled points, process the points to identify and combine tie points using structure-from-motion.

Abstract: A SIMD processing unit processes a plurality of tasks which each include up to a predetermined maximum number of work items. The work items of a task are arranged for executing a common sequence of instructions on respective data items. The data items are arranged into blocks, with some of the blocks including at least one invalid data item. Work items which relate to invalid data items are invalid work items. The SIMD processing unit comprises a group of processing lanes configured to execute instructions of work items of a particular task over a plurality of processing cycles. A control module assembles work items into the tasks based on the validity of the work items, so that invalid work items of the particular task are temporally aligned across the processing lanes. In this way the number of wasted processing slots due to invalid work items may be reduced.

Abstract: It is possible to receive a point cloud as input and perform shape completion with high accuracy. A shape completion unit inputs an input point cloud and a class identification feature output by a class identification unit to a generator that is learned in advance and generates a shape completion point cloud that is to complete a point cloud and is a set of three-dimensional points by receiving, as input, the point cloud and the class identification feature, gaining an integration result obtained by integrating a global feature that is a global feature based on local features extracted from respective points of the point cloud with the class identification feature, and convoluting the integration result, and outputs the shape completion point cloud that is to complete the input point cloud.

Abstract: Described examples include an integrated circuit having depth fusion engine circuitry configured to receive stereoscopic image data and, in response to the received stereoscopic image data, generate at least: first and second focal perspective images for viewing by a first eye at multiple focal distances; and third and fourth focal perspective images for viewing by a second eye at multiple focal distances. The integrated circuit further includes display driver circuitry coupled to the depth fusion engine circuitry and configured to drive a display device for displaying at least the first, second, third and fourth focal perspective images.

Abstract: A method and system for generating a virtual representation of a physical scene, including receiving scene data corresponding to the physical scene, processing the scene data to determine scene components and scene priors corresponding to the scene components, generating, by a plurality of neural networks, dense geometric representations based at least in part on the scene priors, where each dense geometric representation corresponds to a scene component in the scene components, generating a virtual model of the physical scene based at least in part on the dense geometric representations, and generating a virtual representation of the physical scene based at least in part on the scene data, the virtual representation being aligned with the virtual model.

Abstract: The embodiments herein provide a method and a device for the acquisition of three-dimensional information of body parts. The method includes a step of processing an image of an object being exposed to a lighting condition or a sequence of lighting conditions adapted to extract at least two types of information selectable from a group. The group includes a silhouette of the object, a color of the object, a surface orientation of the object, and multi-spectral information of the object.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for scalable three-dimensional (3-D) object recognition in a cross reality system. One of the methods includes maintaining object data specifying objects that have been recognized in a scene. A stream of input images of the scene is received, including a stream of color images and a stream of depth images. A color image is provided as input to an object recognition system. A recognition output that identifies a respective object mask for each object in the color image is received. A synchronization system determines a corresponding depth image for the color image. A 3-D bounding box generation system determines a respective 3-D bounding box for each object that has been recognized in the color image. Data specifying one or more 3-D bounding boxes is received as output from the 3-D bounding box generation system.

Abstract: To improve user convenience as to adjustment for vibration correction of a captured image captured by an image capturing apparatus, a feature portion evaluation unit refers to feature portions selected by a feature portion selection unit and determines whether feature portions necessary for vibration isolation have been acquired from a reference image. The feature portion evaluation unit has the function of notifying information about the feature portions of the reference image in a case where the acquired feature portions do not satisfy a predetermined condition, that is, in a case where the reliability of the acquired feature portions does not reach a predetermined level.

Abstract: A medical image processing apparatus according to an embodiment includes processing circuitry. The processing circuitry is configured to extract a degree of a disease related to the heart from a medical image; and display, when the degree of the disease related to the heart is high, a first index value related to a blood vessel and calculated from one of blood pressure and a blood flow and configured to display, when the degree of the disease related to the heart is low, wall shear stress serving as a second index value related to the blood vessel, as information related to the blood flow of the blood vessel and calculated on the basis of the medical image.

Abstract: Skinning parameters used to animate a virtual avatar can include mesh weights and joint transforms of a skeleton. Systems and methods are provided for determining skinning parameters using an optimization process subject to constraints based on human-understandable or anatomically-motivated relationships among skeletal joints. Input to the optimization process can include a high-order skeleton and the applied constraints can dynamically change during the optimization. The skinning parameters can be used in linear blend skinning (LBS) applications in augmented reality.

Abstract: Embodiments of the systems and methods described herein provide a dynamic animation generation system that can apply a real-life video clip with a character in motion to a first neural network to receive rough motion data, such as pose information, for each of the frames of the video clip, and overlay the pose information on top of the video clip to generate a modified video clip. The system can identify a sliding window that includes a current frame, past frames, and future frames of the modified video clip, and apply the modified video clip to a second neural network to predict a next frame. The dynamic animation generation system can then move the sliding window to the next frame while including the predicted next frame, and apply the new sliding window to the second neural network to predict the following frame to the next frame.

Abstract: A method and apparatus for rendering a computer-generated image using a stencil buffer is described. The method divides an arbitrary closed polygonal contour into first and higher level primitives, where first level primitives correspond to contiguous vertices in the arbitrary closed polygonal contour and higher level primitives correspond to the end vertices of consecutive primitives of the immediately preceding primitive level. The method reduces the level of overdraw when rendering the arbitrary polygonal contour using a stencil buffer compared to other image space methods. A method of producing the primitives in an interleaved order, with second and higher level primitives being produced before the final first level primitives of the contour, is described which improves cache hit rate by reusing more vertices between primitives as they are produced.

Abstract: A computer-implemented method including receiving an input image at a first image stage and receiving a request to generate a plurality of variations of the input image at a second image stage. The method including generating, using an auto-regressive generative deep learning model, the plurality of variations of the input image at the second image stage and outputting the plurality of variations of the input image at the second image stage.

Abstract: Systems, methods, and non-transitory computer readable media are disclosed that include operations to generate a virtual representation of a physical location with spatially localized information related to elements within the location being embedded in the virtual representation. The operations includes receiving description data (e.g., a plurality of images and videos) of the location, the description data being generated via at least one of a camera, a user interface; receive metadata associated elements within the location; generating (e.g., offline or in real-time), via a machine learning model and/or a geometric model, a 3-dimensional (3D) model of the location and elements therein; and generating, based on the 3D model of the location, an information-rich virtual representation of the location by annotating the 3D model with spatially localized metadata associated with the elements within the location and semantic information of the elements.

Abstract: A mobile device executes an augmented reality (AR) software application that detects an orientation of a client device. The AR software application projects a line from a reference position on the client device to a physical object. The AR software application identifies a first location on the physical object that intersects with the line. The AR software application determines an x-coordinate and a y-coordinate of a portion of the physical object included in an image displayed on the client device based on the first location. The AR software application receives a z-coordinate of the portion of the physical object. In response to receiving user input via the client device, the AR software application anchors an augmented reality object at a second location that corresponds to the x-coordinate, the y-coordinate, and the z-coordinate. The orientation of the augmented reality object corresponds to the orientation of the client device.

Abstract: Method for automatically assigning numeric labels to teeth in a dental arch map provides a segmented 3D surface image of patient dentition having identified individual teeth and determined tooth position in the dental arch. A tooth contour for the identified teeth is defined. The tooth type is identified as one of incisor, canine, premolar, or molar. A display of the tooth contour is labeled with an assigned tooth number according to the tooth position, tooth contour, and tooth type. The labeled mapping is displayed.

Abstract: The present invention concerns a swimming goggle including at least one main body including: (a) an outer lens; (b) an inner lens; (c) a sealed gap separating the outer lens and the inner lens; and (d) an inner space arranged adjacent to the inner lens.

Abstract: Systems and methods for image editing are described. Embodiments of the present disclosure provide an image editing system for performing image object replacement or image region replacement (e.g., an image editing system for replacing an object or region of an image with an object or region from another image). For example, the image editing system may replace a sky portion of an image with a more desirable sky portion from a different replacement image. According to some embodiments described herein, thumbnails, region location information, and image metadata from multiple preset images can be stored together and loaded for presentation and selection of a preset image for replacing a region of an image. Once a preset image (e.g., an image with a replacement sky) is selected, a high-resolution version of the image can be loaded and used to generate a composite image.

Abstract: An example method to render an image includes obtaining two-dimensional image information of the image including a target and a scene and depth information associated with a first set of pixels and a second set of pixels of the image. The first set of pixels correspond to the target and the second set of pixels correspond to the scene. In response to depth information associated with a pixel of the image being in the predetermined range, the example method includes associating a first depth index with the pixel. Otherwise, the example method associates a second depth index with the pixel. The example method further includes calculating a weighted average picture level of the image based on grayscale values of pixels of the image, the first depth index and/or the second depth index and rendering the image based on the weighted average picture level.