Abstract: Systems and methods for generating a bias lighting effect are provided. A computer-implemented method can include obtaining a video comprising a plurality of video frames. For each of one or more video frames of the plurality of video frames, the method can include sampling an edge portion of the video frame. The edge portion can include a portion of the video frame adjacent to an edge of the video frame. The method can further include generating a bias lighting effect for the video frame. Generating the bias lighting effect can include inverting the edge portion across the edge and blurring the edge portion. The method can further include displaying the video frame concurrently with the bias lighting effect for the video frame. The bias lighting effect can be displayed adjacent to the edge of the video frame.

Abstract: Some example embodiments relate to a super resolution scanning electron microscope (SEM) image implementing device and/or a method thereof. Provided a super resolution scanning electron microscope (SEM) image implementing device comprising a processor configured to crop a low resolution SEM image to generate a first cropped image and a second cropped image, to upscale the first cropped image and the second cropped image to generate a first upscaled image and a second upscaled image, and to cancel noise from the first upscaled image and the second upscaled image to generate a first noise canceled image and a second noise canceled image.

Abstract: An apparatus and a method of predicting a video frame are provided. The apparatus includes a level encoder configured to extract and learn at least one feature from a video frame, a feature learning unit configured to learn based on the at least one feature or transmit predicted feature data corresponding to the at least one feature, and a level decoder configured to obtain and learn a predicted video frame based on the predicted feature data.

Abstract: According to the method and the apparatus for acquiring a CBCT image based on adaptive sampling according to the exemplary embodiment of the present disclosure, a final CBCT image is acquired by reconstructing a plurality of cone beam computed tomography (CBCT) images acquired based on adaptive sampling so that a dose applied to the target patient may be reduced.

Abstract: A method includes obtaining a first image, determining a first target curve of a first region occupied by the first image, where an area of the first region is equal to an area of the first image, determining a target object on the first image, moving the first image to obtain a second image, where a second target curve of a second region in which the target object on the second image is located is close to or overlaps the first target curve, and the second region is an external polygon of the target object, and cropping out an image part that is of the second image and that is moved out of the first region.

Abstract: A method for analyzing a digestive tract of a patient using ultrasound. The method includes receiving a plurality of ultrasound images taken over a time series, determining movement characteristics of objects within the ultrasound images based on movement of one or more pixels in the ultrasound images over the time series, and outputting a visual data that describes the movement characteristics.

Abstract: A video shooting method includes: enabling a camera application on a terminal and in a slow-motion video recording mode of the camera application, when a motion detection function of the camera application is enabled, starting to capture video frames after a video recording instruction input by a user is detected, recording a second video frame set at a first frame rate automatically when a motion of a human body in a captured first video frame set is a preset motion, and generating a target video, wherein the target video includes a slow-motion video clip, and wherein the slow-motion video clip includes video frames in the second video frame set.

Abstract: A drive assist apparatus includes a storage unit that stores a three-dimensional model indicating a moving region, an input unit that receives, from a sensor group installed in the moving region, first height information indicating a first height which is a height of the mobile object and second height information indicating a second height which is a height of an object that satisfies a predetermined distance criterion from the mobile object, an extraction unit that extracts, from the three-dimensional model, a first plan view based on the first height information and a second plan view based on the second height information, a generation unit that generates a combined map for two-dimensionally showing the moving region and assisting the driving of the mobile object by combining the first plan view and the second plan view, and an output unit that transmits a generated combined map to the mobile object.

Abstract: A method for recommending a product, includes: providing product information having a product description, a main image, and a plurality of alternate images; for each alternate image: performing saliency segmentation to obtain a segment and a background, defining a segment bounding box enclosing the segment, extracting image feature from the segment bounding box, extracting textual feature from the product description, extracting main image feature from the main image, matching the image feature of the segment bounding box to the textual feature and the main image feature, and determining the alternate image as a qualified alternate image if the image feature of the segment bounding box matches the textual feature and the main image feature; and when a number of the qualified alternate image equals to or is greater than a threshold number, recommending the product to customers.

Abstract: Systems and methods for image captioning are described. One or more aspects of the systems and methods include generating a training caption for a training image using an image captioning network; encoding the training caption using a multi-modal encoder to obtain an encoded training caption; encoding the training image using the multi-modal encoder to obtain an encoded training image; computing a reward function based on the encoded training caption and the encoded training image; and updating parameters of the image captioning network based on the reward function.

Abstract: A computer-implemented method is for classifying a region of interest in a medical image data set depicting a body part of a patient. The region of interest contains an object of potential pathological relevance. In an embodiment, the method includes a plurality of steps. One step is directed to generate a plurality of different representations of the region of interest. Another step is directed to determine, for each of the representations, a classification to generate a corresponding plurality of classifications. Thereby, each classification indicates a pathological relevance of the object in the respective representation. Yet, a further step is directed to calculate an ensemble classification for the region of interest based on the plurality of classifications.

Abstract: A computer-implemented method of training an image analysis model. The image analysis model comprises a coupling layer that determines an output vector of discrete values from an input vector of discrete values. First, a machine learnable submodel of the coupling layer is trained to predict a second input part of the coupling layer from a first input part of the coupling layer. Next, the image analysis model is trained. This involves applying the coupling layer by applying the machine learnable submodel to the first input part to obtain a prediction of the second input part; and determining a second output part by applying an invertible mapping to the second input part defined by the prediction of the second input part. The mapping maps a predicted value of an element of the second input part to a fixed value independent from the predicted value.

Abstract: A registration means for storing an image of a product as a registration image in association with information representing the passing sequence that the product passed through an upstream side process; a management means for managing the matching sequence in a downstream side process; and a matching means for performing matching between an image of a product carried into the downstream side process and the registration image according to the matching sequence, are included. Each time the matching means succeeds in matching, the management means updates the matching sequence to sequence in which registration images not having succeeded in matching with any matching image are put in order on the basis of the passing sequence that the products passed through the upstream side process.

Abstract: The devices, systems, and methods can improve magnetic resonance imaging (MRI), MR spectroscopy (MRS), MR spectroscopic imaging (MRSI) measurement(s), thereby providing more reliable quantification. The method may include a method for correcting MR image(s)/spectrum. The method may include providing an inhomogeneity field/response map of a region of interest; and providing MR image(s)/spectrum of the region of interest. The method may include determining an intravoxel/voxel inhomogeneity correction coefficient for each voxel of at least one subregion of the region of the interest using the inhomogeneity field/response map. The method may include correcting each voxel of the MR image(s)/spectrum of the region of interest using the intravoxel/voxel inhomogeneity correction coefficient.

Abstract: The invention disclose a method, an apparatus, a device and a storage medium for detecting a card surface picture. The method comprises the following steps of: identifying region information of an image to be shown in a target picture; generating a picture to be detected according to the region information of the image to be shown; synthesizing the picture to be detected and a preset picture to obtain a synthesized picture; and in response to that it is detected that a fourth pixel value is contained in the synthesized picture, determining that the target picture is unqualified, which improves the efficiency of picture review and further improves the efficiency of card fabrication.

Abstract: A substrate inspection method includes: (i) acquiring a plurality of defect of interest (DOI) images of a substrate having a DOI, under a corresponding plurality of different optical conditions, (ii) acquiring a plurality of DOI difference images from differences between the plurality of DOI images and a reference image, and (iii) acquiring a plurality of DOI difference-of-difference (DOD) images from differences between the plurality of DOI difference images. The method also includes setting two optical conditions corresponding to a DOI DOD image having the highest signal-to-noise ratio (SNR) among the plurality of DOI DOD images, as a first optical condition and a second optical condition, and acquiring a first image of the substrate under the first optical condition and a second image of the substrate under the second optical condition.

Abstract: The invention describes a generic framework for hand-eye calibration of camera-guided apparatuses, wherein the rigid 3D transformation between the apparatus and the camera must be determined. An example of such an apparatus is a camera-guided robot.

Abstract: In an implementation, a method for pixel level roadway crack segmentation is provided. The method includes: receiving a plurality of roadway range images; generating a plurality of image patches from the plurality of roadway range images; generating a crack map for the plurality of image patches by a DCNN; and generating a crack map for the plurality of roadway range images based on the generated crack map for the plurality of image patches.

Abstract: System, apparatus and method of image processing to detect a substance spill on a solid surface such as a floor is disclosed. First data representing a first image, captured by an image sensor, of a region including a solid surface, is received. A trained semantic segmentation neural network is applied to the first image data to determine, for each pixel of the first image, a spill classification value associated with the pixel, the determined spill classification value for a given pixel indicating the extent to which the trained semantic segmentation neural network estimates, based on its training, that the given pixel illustrates a substance spill. The presence of a substance spill on the solid surface is detected based on the determined spill classification values of the pixels of the first image.

Abstract: Systems, methods, and data storage devices for image grouping in an end user device using trained machine learning group classifiers are described. The end user device may include an image group classifier configured to classify new image data objects using an image classification algorithm and set of machine learning parameters previously trained for a specific image group. The end user device may determine embeddings that quantify features of the target image object and use those embeddings and the image group classifier to selectively associate group identifiers with each new image data object received or generated by the end user device. Calibration, including selection and training, of the image group classifiers and ranking of classified images are also described.