Abstract: A processor-implemented method includes: identifying input components of a semiconductor pattern of an original input image from the original input image corresponding to an application target of a process for manufacturing a semiconductor, generating an augmented input image by transforming a transformation target comprising one or more of the input components from the original input image; and executing a neural model for estimating pattern transformation according to the process based on the augmented input image.

Abstract: Various implementations disclosed herein include devices, systems, and methods that generates values for a representation of a face of a user. For example, an example process may include obtaining sensor data (e.g., live data) of a user, wherein the sensor data is associated with a point in time, generating a set of values representing the user based on the sensor data, and providing the set of values, where a depiction of the user at the point in time is displayed based on the set of values. In some implementations, the set of values includes depth values that define three-dimensional (3D) positions of portions of the user relative to multiple 3D positions of points of a projected surface and appearance values (e.g., color, texture, opacity, etc.) that define appearances of the portions of the user.

Abstract: The following relates generally to using augmented reality (AR) for inspection of a property. In some examples, underlay layer data may be received (e.g., from a camera of an AR viewer device); and overlay layer data may be received (e.g., from a different camera or other overlay layer device). An AR display may be created by correlating the underlay layer data with the overlay layer data. A property inspection indicia may be identified based upon the correlated overlay layer data.

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods for embedding perturbation data via a machine learning model and filtering, aligning, and aggregating the embeddings to generate a genome-wide perturbation database for real-time generation of perturbation heatmaps. In particular, in one or more embodiments, the disclosed systems can receive a plurality of perturbation images portraying cells from a plurality of wells corresponding to a plurality of cell perturbations. Further, the systems can generate, utilizing a machine learning model, a plurality of well-level image embeddings from the plurality of perturbation images. Moreover, the systems can align, utilizing an alignment model, the plurality of well-level image embeddings to generate aligned well-level image embeddings. Additionally, the systems can aggregate, according to perturbations of one or more perturbation experiments, the well-level image embeddings to generate perturbation-level image embeddings.

Abstract: Disclosed is a data processing method including obtaining deviation information indicating a degree of deviation of target data from reference data, and displaying, on the target data, information corresponding to the deviation information, wherein the obtaining of the deviation information includes obtaining the deviation information by comparing data other than data that deteriorates reliability of the deviation information from among the target data with the reference data.

Abstract: An image processing method includes acquiring an image in which a cell structure having a stained vascular network structure is imaged, applying a wavelet transform to the image such that a contour image in which contours of the vascular network structure are extracted is generated, and repeatedly excluding object pixels from object boundaries recognized in the contour image such that a skeleton image in which skeletons having a line width of a predetermined number of pixels are extracted is generated.

Abstract: An electronic device for gesture recognition on resource-constrained devices is provided. The electronic device controls storage of a plurality of first consecutive image frames in a first buffer of a first length. The plurality of first consecutive image frames corresponds to the first length. The electronic device recognizes a first hand sign of a plurality of hand signs in a first subset of image frames of the plurality of first consecutive image frames. The electronic device controls storage of the recognized first hand sign in a second buffer of a second length based on the determination that a ratio of a number of the first subset of image frames and the first length is one of equal to or greater than the threshold. The electronic device determines a gesture corresponding to one or more hand signs of the plurality of hand signs stored in the second buffer.

Abstract: Embodiments of present disclosure relates to method and dimension determination system for determining dimensions of carton box. The dimension determination system receives plurality of datapoints associated with open carton box. The dimension determination system extracts depth datapoints from plurality of datapoints for identifying plurality of flaps. The dimension determination system determines height of open carton box using depth datapoints. The dimension determination system generates contour and estimates rectangle figure for contour of open carton box. Further, the dimension determination system determines width and length by extracting vertices of rectangle figure. Thereafter, the dimension determination system utilises width, length and height to obtain dimensions of open carton box. Thus, the present disclosure automatically determines dimensions of carton box.

Abstract: Systems, apparatus, and methods for super-resolution of non-uniform spatial blur. Non-uniform spatial blur presents unique challenges for conventional neural network processing. Existing implementations attempt to handle super-resolution with a “brute force” optimization. Various embodiments of the present disclosure subdivide the super-resolution function into sub-steps. “Unfolding” super-resolution into smaller closed-form functions allows for operation using generic plug-and-play convolutional neural network (CNN) logic. Additionally, each step can be optimized with its own step-specific hyper parameters to improve performance.

Abstract: An image processing device and method for converting an input image having a first dynamic range into an output image having a second dynamic range higher than the first dynamic range are provided. The image processing device includes a mapping unit for transforming an input luminance value associated with a pixel of the input image into an output luminance value associated with the corresponding pixel in the output image. The mapping unit is configured to apply a continuous and increasing function mapping a first range of values into a second range of values, whereby the highest value in the first range of values is mapped by the continuous and increasing function into a supremum of the second range of values. The mapping unit is configured to adjust the supremum depending on a peak luminance.

Abstract: A method of analyzing one or three dimensional ejecta, including depth information, volume, mass and kinematic properties, resulting from a dynamic event. The method comprising using a single trigger, single image sensor three dimensional photography system comprising a first plurality of pulsed laser beams, a stage for the dynamic event to occur and be observed and where the first plurality of pulsed laser beams converge. The laser beams are recombined into a single pulsed laser beam after the stage for capture on the single image sensor to a subframe for each pulsed laser beam, the subframes being combined into a unique frame for each time step. Each static frame becomes a frame in a video of the event where dT is controlled by the laser repetition rate. The frames are analyzed with respect to time to determine the displacement, velocity and/or acceleration thereof. Plural N×N×M arrays of N×M pixels taken from the image sensor are combined to create a three dimensional image of each desired ejectum.

Abstract: The present invention relates to the field of image processing. Disclosed are a neural network-based pose estimation and registration method for heterogeneous images. In the present invention, a phase correlation algorithm is optimized to be differentiable and embedded into an end-to-end learning network framework, and a neural network-based pose estimation method for heterogeneous images is constructed. According to the method, an optimal feature extractor can be found for a result of image matching, a solution can be obtained without exhaustive evaluation, and good interpretability and generalization capability are achieved. The test results show that the present invention allows for accurate pose estimation and registration for heterogeneous images and shortening of the required time, has high accuracy and real-time performance, can meet actual application requirements, and can be applied in fields such as robot self-positioning.

Abstract: Some embodiments of the disclosure provide a measuring method and measuring apparatus of a blood vessel diameter of a fundus image. In some examples, the measuring method include the following steps: acquiring a fundus image, generating a blood vessel segmentation image based on the fundus image, performing resolution enhancement on the blood vessel segmentation image, extracting blood vessel skeletons from the enhanced blood vessel segmentation image and performing fitting on the blood vessel skeletons to obtain a continuous blood vessel skeleton and a vessel diameter measurement direction of measurement pixel points, generating a blood vessel contour corresponding to the measurement pixel points based on the enhanced blood vessel segmentation image, and calculating a blood vessel diameter corresponding to the measurement pixel points based on a number of blood vessel pixel points in the blood vessel contour.

Abstract: Methods, computer systems, and apparatus, including computer programs encoded on computer storage media, for processing sensor data. In one aspect, a method includes obtaining image data representing a camera sensor measurement of a scene; obtaining radar data representing a radar sensor measurement of the scene; generating a feature representation of the image data; generating a respective initial depth estimate for each of a subset of the plurality of pixels; generating a feature representation of the radar data; for each of the subset of the plurality of pixels, generating a respective adjusted depth estimate for the pixel using the initial depth estimate for the pixel and the radar feature vectors for a corresponding subset of the plurality of radar reflection points; generating a fused point cloud that includes a plurality of three-dimensional data points; and processing the fused point cloud to generate an output that characterizes the scene.

Abstract: In one embodiment, a method includes accessing an image of one or more target buildings captured by a device; detecting target building planes of the one or more target buildings in the image using a machine learning program; for each of the target building planes: generating a target embedding and target feature points for the target building plane; querying a three-dimensional (3D) map using the target embedding to identify at least one matching building plane for the target building plane; using the target feature points to identify matching feature points of the at least one matching building plane; and determining three-dimensional (3D) locations of the matching feature points using the 3D map; and determining a pose of the device by comparing the target feature points of the target building planes and the 3D locations of the matching feature points associated with the target building planes.

Abstract: In a method of classifying defects of a pattern, SEM images having different colors on the same pattern are aligned and merged. The defects of the pattern are located by color difference in the merged SEM image. The defects of the pattern are firstly located at the location of the defects. A design in which a layout of the pattern is drawn, and the merged SEM image are aligned. The defects of the pattern are secondly classified by comparing the layout of the pattern drawn in the design and the merged SEM image. A type of the defects of the pattern is determined by combination of the first and second classifications of the defects of the pattern.

Abstract: A method receives first and second data generated from a first and second domains including first and second set of objects, receiving first class labels for each of the first set of objects, and receiving second class labels for each of the second set of objects. The method generates a training dataset by augmenting the first data and corresponding first class labels, and locally updating neural network parameters of a model based on the training dataset. The method generates a validation dataset by augmenting the second data and corresponding second class labels, and globally updating the neural network parameters of the model based on the validation dataset. The method also generates multiple target labels for target data generated from a target domain including a third set of objects after globally updating the neural network parameters of the model based on the validation dataset.

Abstract: The present invention provides an evaluation method for more precisely determining the blur correction capability of an image capturing apparatus. The method for evaluating the blur correction effect of the image capturing apparatus comprises acquiring a first change amount for each image capture from a predetermined captured number of subject images that have been captured by an image capturing apparatus that has been vibrated; dividing the predetermined number of captured images into a plurality of segments, and calculating a second change amount for each segment by performing statistical processing on the first change amounts in each of the plurality of segments; determining the largest second change amount from among the second change amounts in each segment to serve as a third change amount; and performing evaluation of the blur correction effect based on the third change amount.

Abstract: An image foreground-background segmentation method and system based on sparse decomposition and graph Laplacian regularization are disclosed. Firstly, an image is divided into a plurality of non-overlapping image blocks; Then, a foreground-background segmentation model of the image is established according to the image blocks; An image segmentation problem is divided into several sub-problems, which are solved by iteration; Finally, after the iteration, solutions of the problem are obtained; The obtained solutions are respectively matrixed and patched to obtain a foreground image, which is a foreground image of the whole image. The image foreground-background segmentation method uses the linear combination of graph Fourier basis functions to better represent the smooth background region. In addition, the graph Laplacian regularization is used to characterize the connectivity of foreground text and graphics while keeping sharp foreground text and graphics contours.

Abstract: A method for generating a visual feature map includes receiving a first image, detecting a first set of keypoints from the first image, extracting a visual feature descriptor of each of the first set of keypoints, receiving a first set of 3D point cloud data associated with a location where the first image is captured, and determining a 3D coordinate value of each of the first set of keypoints using the first set of 3D point cloud data.