Abstract: Methods and systems for image analysis are provided, and in particular for identifying a set of base-calling locations in a flow cell for DNA sequencing. These include capturing flow cell images after each sequencing step performed on the flow cell, and identifying candidate cluster centers in at least one of the flow cell images. Intensities are determined for each candidate cluster center in a set of flow cell images. Purities are determined for each candidate cluster center based on the intensities. Each candidate cluster center with a purity greater than the purity of the surrounding candidate cluster centers within a distance threshold is added to a template set of base-calling locations.

Abstract: A method for training an eye tracking model is disclosed, as well as a corresponding system and storage medium. The eye tracking model is adapted to predict eye tracking data based on sensor data from a first eye tracking sensor. The method comprises receiving sensor data obtained by the first eye tracking sensor at a time instance and receiving reference eye tracking data for the time instance generated by an eye tracking system comprising a second eye tracking sensor. The reference eye tracking data is generated by the eye tracking system based on sensor data obtained by the second eye tracking sensor at the time instance. The method comprises training the eye tracking model based on the sensor data obtained by the first eye tracking sensor at the time instance and the generated reference eye tracking data.

Abstract: Disclosed herein are embodiments related to robot-aided product inspection. For example, an inspection apparatus may include a processing device to communicatively couple to a robotic apparatus, to a display device, and to a user input device. The processing device may: receive an image of an item, wherein the image was captured by the robotic apparatus; generate a proposed classification for the image, wherein the classification indicates an attribute of the item based on the image; cause the image to be displayed on the display device along with the proposed classification; and receive an indication from the user input device of a final classification of the image.

Abstract: A method of detecting a target includes generating an image pyramid based on an image on which a detection is to be performed; classifying candidate areas in the image pyramid using a cascade neural network; and determining a target area corresponding to a target included in the image based on the plurality of candidate areas, wherein the cascade neural network includes a plurality of neural networks, and at least one neural network among the neural networks includes parallel sub-neural networks.

Abstract: Embodiments of the present disclosure disclose an object detection method, an object detection apparatus, a related device, a storage medium and a vehicle. The method includes the following. Scene point cloud data is inputted into a point cloud feature extraction model pretrained to obtain a point cloud descriptive feature of the scene. Scene image data is input into an image feature extraction model pretrained to obtain an image descriptive feature of the scene. The point cloud descriptive feature and the image descriptive feature of each target region in the scene are input into an object identifying model to obtain a detection result of an object included in each target region.

Abstract: A method of performing visualized measurement on thickness distribution of a paint film and an apparatus therefor. A measurement target region is heated by a heating unit that applies a light beam while moving relative to the measurement target region of a measurement target structure. A sensing unit moving together with the heating unit generates a plurality of thermal images related to a phenomenon in which thermal energy is propagated in the measurement target region by scanning and photographing the heated measurement target region. The thermal images in a dynamic state are converted into time-spatial-integrated thermal images in a static state by performing coordinate transformation according to a time-spatial-integrated coordinate transformation algorithm. A thickness of the paint film is calculated by using a Fourier thermal conduction equation.

Abstract: Embodiments of the present invention provide systems, methods, and computer storage media for one-shot and few-shot image segmentation on classes of objects that were not represented during training. In some embodiments, a dual prediction scheme may be applied in which query and support masks are jointly predicted using a shared decoder, which aids in similarity propagation between the query and support features. Additionally or alternatively, foreground and background attentive fusion may be applied to utilize cues from foreground and background feature similarities between the query and support images. Finally, to prevent overfitting on class-conditional similarities across training classes, input channel averaging may be applied for the query image during training. Accordingly, the techniques described herein may be used to achieve state-of-the-art performance for both one-shot and few-shot segmentation tasks.

Abstract: A method for identifying one or more imaging properties. The method may comprise identifying one or more candidate mud constants, taking one or more measurements from a borehole with a downhole tool to form an image log, inputting into a machine learning model one or more inputs such that the machine learning model outputs one or more estimated imaging properties, and inputting into a forward model at least the one or more estimated imaging properties and the one or more candidate mud constants such that the forward model outputs one or more reconstructed tool responses. The method may further comprise computing a misfit between the one or more reconstructed tool responses and the one or more measurements, picking a mud candidate from the one or more candidate mud constants based at least in part on the misfit, and producing one or more imaging properties from the picked mud candidate.

Abstract: A method for analyzing media to evaluate talent of a subject portrayed in the media is provided. The steps describe receiving a plurality of the media at a server, the media being performance based media; transforming the media into at least a data set corresponding to at least a characteristic of the media at the server, wherein the server is in communication with a node; identifying features in the data set that correspond to a probability of success of the subject in the media; receiving subjective opinion polling data from a plurality of participants as to whether the subject in the media has talent; training the node via machine learning algorithm to identify, using the identified features in the data set and the subjective opinion polling data provided, whether the subject is likely to have talent; and executing the machine learning algorithm to output a score relating to the talent of the subject in the media. A system for discovering talent is also provided herein.

Abstract: An image processing apparatus comprises a first obtaining unit configured to obtain a range of a transmittance of air that is designated in accordance with a user operation to a captured image, and a generation unit configured to generate a corrected image in which influence of a microparticle component in the air on the captured image is suppressed based on the range.

Abstract: An information processing system creates a teacher database configured to train an analysis model from an observation image and labeling information corresponding to the observation image using an information processor. This system includes a storage unit, an image processing unit, and a teacher database creating unit. The storage unit stores image processing data formed of information showing a relationship between an observation condition and a parameter relating to the observation image. Further stores a first observation image, a first observation condition, and first labeling information. The image processing unit accepts the first observation image and the first observation condition as inputs, performs image processing corresponding to the parameter to the first observation image based on the image processing data, and creates a second observation image corresponding to a second observation condition.

Abstract: An information processing apparatus includes a memory; and a processor configured to determine a plurality of initial image processing programs and to add the initial image processing programs to an image processing program group, extract at least two image processing programs from the image processing program group, generate a candidate of a next-generation image processing program from the extracted image processing programs, based on genetic programming, calculate a fitness of the candidate of the next-generation image processing program, using a learning data item including an input image and a target processing result, and determine the next-generation image processing program or update the image processing program group based on the calculated fitness, wherein the processor is configured to determine, as some of the initial image processing programs, at least some of first image processing programs included in the image processing program group upon previous determination of a next-generation image proc

Abstract: Disclosed herein are cell differentiating systems for differentiating cells. The systems comprise an input means for receiving a reconstructed image of a cell based on a holographic image of the cell in suspension, and a cell recognition means for determining cell characterization features from the reconstructed image of the cell for characterization of the cell. The cell recognition means thereby is configured for determining, as cell recognition features, image moments.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media for efficiently and flexibly extracting reusable geometric assets from an arbitrary arrangement of vector geometry within a digital image. For example, the disclosed systems can organize vector geometry of a digital image by structuring geometric objects into groups (e.g., clusters). The disclosed systems can assign mnemonics to these groups and transform the digital image into a mnemonic sequence. Moreover, the disclosed systems can utilize various computer-implemented algorithms to identify and filter patterns within the mnemonic sequence. The disclosed systems can then generate pattern scores for these patterns and identify which patterns of geometric objects to include within a set of reusable geometric assets.

Abstract: Partial-image search criteria are derived that include a combination of multiple partial-image user selections of portions of at least one image. Visual search results are returned responsive to a search for images that correspond to the derived partial-image search criteria. The visual search results include at least one identified image that includes features that correspond to the derived partial-image search criteria that include the combination of the multiple partial-image user selections.

Abstract: Automated explanation of machine learning predictions, by: Obtaining an input, a machine learning model, and a classification score produced by application of the machine learning model to the input. Optimizing iterative application of the machine learning model to perturbations of the input, wherein a target of said optimizing is at least a predefined non-zero classification score. Generating, based on one of the perturbations which achieved the predefined non-zero classification score, an explanation to the classification score produced by the application of the machine learning model to the input.

Abstract: An apparatus comprising: an obtaining unit configured to obtain target data as a result of discrimination of each portion of input data performed by a discriminator having learned in advance by using existing labeled data; a setting unit configured to set each portion of the target data, which is effective for additional learning of the discriminator, as local data; a determining unit configured to determine not less than one partial region of the target data, which accepts labeling by a user, based on a distribution of the set local data in the target data; and a display control unit configured to cause a display unit to display the determined not less than one partial region.

Abstract: Systems and methods are disclosed for processing an electronic image corresponding to a specimen. One method for processing the electronic image includes: receiving a target electronic image of a slide corresponding to a target specimen, the target specimen including a tissue sample from a patient, applying a machine learning system to the target electronic image to determine deficiencies associated with the target specimen, the machine learning system having been generated by processing a plurality of training images to predict stain deficiencies and/or predict a needed recut, the training images including images of human tissue and/or images that are algorithmically generated; and based on the deficiencies associated with the target specimen, determining to automatically order an additional slide to be prepared.

Abstract: An image analysis system and method classify cells in a tissue image. The system and method may extract at least one image feature characterizing an object in the tissue image. Based on the extracted image feature, cells may be classified according to at least one predefined membrane pattern. For each classified cell, a class label that identifies a class to which the classified cell belongs may be outputted.

Abstract: A method for controlling, by a controller, an electronic device is provided. The method may include: sensing, by a sensor, a first radiation signal incident on the sensor; generating, by the sensor, a first output signal based on the first radiation signal; recognizing a human body based on the first output signal; determining a position of the human body as being located in one of an indoor space or an outdoor space based on the first output signal; and generating a control signal for controlling the electronic device connected via a wired or wireless network based on the position of the human body. A learning model includes a deep neural network generated through machine learning and transmission of a control signal may be performed in an Internet of Things (IoT) environment using a 5G network.