Abstract: An automated adjustable mannequin is provided for use as a platform for producing computer generated fashion model clothing displays. The clothing displays may be static photographs or video. The mannequin combines manually adjustable dress forms currently used by seamstresses and clothing designers with three dimensional scanning and photographic technologies. Actual body scan data of a living model may be used to automatically adjust the shape or contours of a dress form in real time to match the size and dimensions of a previously photographed model. The automated adjustments to the mannequin may be made at a rate of one frame at a time or 30 frames per second, where the body-form adapts to changes in the underlying model's movements.

Abstract: Provided are a picture enhancement method and apparatus. The picture enhancement method includes: for any one pixel, which meets a first filtering condition, in a current block, determining a first pixel value of the pixel after first filtering; and performing enhancement on a pixel value of the pixel based on the first pixel value and a second pixel value of the pixel point before the first filtering, so as to obtain a third pixel value, which has been subjected to enhancement, of the pixel.

Abstract: A system determines an input video and a first annotated image from the input video which identifies an object of interest. The system initiates a tracker based on the first annotated image and the input video. The tracker generates, based on the first annotated image and the input video, information including: a sliding window for false positives; a first set of unlabeled images from the input video; and at least two images with corresponding labeled states. A semi-supervised classifier classifies, based on the information, the first set of unlabeled images from the input video. If a first unlabeled image is classified as a false positive, the system reinitiates the tracker based on a second annotated image occurring in a frame prior to a frame with the false positive. The system generates an output video comprising the input video displayed with tracking on the object of interest.

Abstract: A system and method to automatically extract bitting information, such as keycode information, for replicating original/master key. System comprises an imaging device that scans the key blade of the original key and sends images to the server where the keycode gets extracted using an iterative software algorithm based on maximum likelihood principle. In one example a system comprises a kiosk which includes SBC (Single Board Computer), imaging device, touch screen and transponder identification module determines compatible SKU (from internal SKU database) if needed, and a compatible key with a matched cut key blade gets delivered to the location of Kiosk to get paired to the car. Another type system comprises a Smartphone/tablet with CKE application instead of KIOSK to scan the original keyblade. The application sends images to a central server for decoding and cutting a duplicate key, and a compatible duplicate key with matched cut key blade is delivered to the associated customer.

Abstract: An apparatus and method for generating a quality diagnostic of analog ECG (electrocardiogram) data. The apparatus includes at least a processor and a memory communicatively connected to the at least a processor, wherein the memory contains instructions configuring the at least a processor to receive ECG data, extract a plurality of ECG parameters from the ECG data, convert the ECG data to one or more digitized ECG signals, validate the digitized ECG signals, wherein validating includes classifying the plurality of digitized ECG signals to a plurality of preliminary ECG parameters and determining an accuracy status of the extracted plurality of ECG parameters by comparing the plurality of preliminary ECG parameters to the extracted plurality of ECG parameters, and generate a quality diagnostic of the ECG data based on validation of the digitized ECG signals.

Abstract: An object detection method, the method includes: determining an object event image from event images collected by an event camera in the case of obtaining an original radar point cloud; obtaining, by projecting the original radar point cloud according to a pre-calibrated extrinsic matrix, relative to the event camera, of a radar, a projected point cloud of the original radar point cloud in a coordinate system of the event camera; determining an object radar point from the original radar point cloud according to a positional relation between projected points in the projected point cloud and an image contour in the object event image; and performing object detection according to the object radar point.

Abstract: Methods for detecting missing or ectopic teeth. At least one tooth that is missing or ectopic may be detected in a dental arch by determining that a gap between adjacent teeth of a digital model of the dental arch exceeds a gap threshold. At least one tooth of the dental arch may be determined to be ectopic by performing an ectopic analysis of a region of the digital model of the dental arch above or below the gap. Anatomical identifiers may be assigned to the teeth of the digital model of the dental arch to account for the at least one tooth that is missing or ectopic. An orthodontic treatment plan to reposition one or more teeth of the dental arch using the assigned anatomical identifiers of the teeth of the digital model of the dental arch may be created.

Abstract: Method and device for detecting and classifying a not-known object by a device having a two-dimensional field of view. The method includes a learning step, a detecting step and an inspection step. The learning step includes positioning a first known object in a field of view of the device and then positioning a second known object in the field of view of the device, to obtain images of the first and second known objects. Distinctive characteristics and common characteristics of the objects are identified and used to obtain an equality descriptor relative to common elements of the first and second known objects and a difference descriptor allowing association of the first and second known objects to two distinct classes. The device is then used to detect a third, not-known object. The equality descriptor and the difference descriptor are then used in an inspection step to classify the third not-known object.

Abstract: A method for enhancing a reference image includes processing the reference image via a generator network to provide an enhanced image by applying a first pre-processing to the reference image and then applying a plurality of first layers of residual blocks to extract features. A first post-processing is then applied. A conversion layer, such as one of a night vision conversion layer and an air pollution conversion layer, is then applied. A plurality of second layers of residual blocks are applied after one of the night vision conversion layer and the air pollution conversion layer are applied, and then a second post-processing is applied which includes upscaling the output of the plurality of second layers of residual blocks, resulting in the enhanced image.

Abstract: The presently disclosed subject matter there includes a computerized device that comprises a processing circuitry, and one or more sensors including at least one image sensor configured to continuously capture images of an environment; the processing circuitry is configured to process the captured images and: detect an object within the environment; track a trajectory of movement of the object; select a best fitting motion model to the trajectory and determine a state of the object based on the best fitting motion model; and determine at least one intersection point between the object and a target object.

Abstract: The present concepts relate to lossless data compression techniques for reducing the size of a data structure. Certain data in the data structure that can be either recovered from another source or rebuilt from other available information may be removed from the data structure. To further reduce data size, the retained data in the data structure may be packed into a smaller-bit encoding data type. Additionally, to reduce the data size even more, the packed data may be zipped using a lossless data compression algorithm. To regain the original data structure, the process may be reversed. The zipped data may be unzipped using a lossless data decompression algorithm. The packed data may be unpacked into the original bit-sized data encoding. The removed data may be restored by either recovering it from another source or rebuilding it from other available information.

Abstract: Disclosed in the embodiments of the present disclosure are a video processing method and apparatus, and an electronic device and a storage medium. The method include: determining a blurred video frame in an initial video; removing the blurred video frame from the initial video, so as to obtain an intermediate video, which dose not comprise the blurred video frame; on the basis of a video frame in the intermediate video that has a timestamp adjacent to a target timestamp, determining a video frame to be inserted, wherein the target timestamp is the timestamp of the blurred video frame; and inserting into a position in the intermediate video that corresponds to the target timestamp, the video frame to be inserted, so as to obtain a target video.

Abstract: Apparatuses, systems, and techniques to generate blue noise masks for real-time image rendering and enhancement. In at least one embodiment, a noise mask is generated and applied to one or more images to generate one or more enhanced images for image processing (e.g., real-time image rendering). In at least one embodiment, the noise mask is able to handle the temporal domain (e.g., add time to the spatial domain) to improve image quality when rendering images over multiple frames.

Abstract: Provided are an apparatus and method for honeycomb artifacts removal for removing honeycomb artifacts in images received from an optical fiber imaging device through a honeycomb artifact removal model. The honeycomb artifact removal apparatus according to an embodiment includes a control unit configured to create a training dataset, build the honeycomb artifact removal model with the created training data, remove a honeycomb artifact in an input image through the built honeycomb artifact removal model and output a corrected image, wherein the control unit is configured to perform preprocessing of a raw image, acquire a honeycomb artifact image through the optical fiber imaging device, and synthesize the preprocessed raw image with the honeycomb artifact image to generate a honeycomb artifact synthesized image, and the training dataset is created by mapping the honeycomb artifact synthesized image as input data and the preprocessed raw image as output data.

Abstract: A surgical simulator for simulating a surgical scenario comprises a display system, a user interface, and a controller. The controller includes one or more processors coupled to memory that stores instructions that when executed cause the system to perform operations. The operations include generating simulated surgical videos, each representative of the surgical scenario. The operations further include associating simulated ground truth data from the simulation with the simulated surgical videos. The ground truth data corresponds to context information of at least one of a simulated surgical instrument, a simulated anatomical region, a simulated surgical task, or a simulated action. The operations further include annotating features of the simulated surgical videos based, at least in part, on the simulated ground truth data for training a machine learning model.

Abstract: Disclosed in the present disclosure are an image-text mutual retrieval model training method and apparatus, an image-text mutual retrieval method, and a device, applied to the technical field of retrieval. The image-text mutual retrieval model training method includes: acquiring training data pairs; inputting the training data pair in an initial model, and extracting text coding features of text training data and image coding features of image training data by using a text coding module and an image coding module in the initial model, respectively; calculating a training loss based on the text coding features and the image coding features, and performing parameter adjustment on the initial model based on the training loss; and in response to the training loss meeting a convergence condition, determining the initial model after the parameter adjustment as an image-text mutual retrieval model.

Abstract: A method of detecting a defect in a stacked structure of a display panel includes collecting a first image of the defect and a plurality of layers in the stacked structure from a database, learning a defect information of the defect and a layer information of the layers using a deep learning model based on the first image and detecting a location of the defect among the layers by the defect information and the layer information.

Abstract: The present invention generally relates to the field of replicating or copying keys. More specifically, the present invention relates to creating a copy of a master key based on a captured image of the master key. The present invention identifies a set of target key information based on vehicle and or lock information as well as the image of the master key to provide for the cutting of a duplicate key blade copy to be delivered to a user or to another location. Additional key information may also be captured along with the image of the master key.

Abstract: The present invention relates to a method of assisting in diagnosis of a target heart disease using a retinal image, the method including: obtaining a target retinal image which is obtained by imaging a retina of a testee; on the basis of the target retinal image, obtaining heart disease diagnosis assistance information of the testee according to the target retinal image, via a heart disease diagnosis assistance neural network model which obtains diagnosis assistance information that is used for diagnosis of the target heart disease according to the retinal image; and outputting the heart disease diagnosis assistance information of the testee.

Abstract: System and method for extracting human pose information from an image, comprising a feature extractor connected to a database, a convolutional neural network (CNN) with a plurality of CNN layers. Said system/method further comprising at least one of the following modules: a 2D body skeleton detector for determining 2D body skeleton information from the human-related image features; a body silhouette detector for determining body silhouette information from the human-related image features; a hand silhouette detector for determining hand silhouette detector from the human-related image features; a hand skeleton detector for determining hand skeleton from the human-related image features; a 3D body skeleton detector for determining 3D body skeleton from the human-related image features; and a facial keypoints detector for determining facial keypoints from the human-related image features.