Abstract: Provided herein are methods of drug design comprising using a machine learning model to classify images of tissue samples as being of tissue samples from subject in a first state and tissue samples from subjects in a second state and extracting from the machine learning model groups of biological components comprising at least two biological components whose spatial relationship contributed to the classifying. Methods of drug design, as well as methods of converting a subject from a first state to a second state are also provided.

Abstract: Disclosed herein are methods, systems, and computer-readable media for generating image captions for training a machine learning model. Current image generation models are hindered by the prevalence of improper or inaccurate captions, which leads to suboptimal training data. This results in less effective image generation models. Disclosed systems and methods involve obtaining a text-to-image dataset including one or more digital image-caption pairs. Systems and methods involve generating a recaptioned dataset by applying an image captioner model to images in the text-to-image dataset. An image captioner model can be trained with an improved image dataset, a first tuning stage, and a second tuning stage, for improved performance.

Abstract: A computer-implemented method comprises: receiving, from a first local site remote from a model aggregator device, a local update of a machine learned model and a parameterization of local data, wherein the local update was generated at the first local site based on the local data; generating a synthetic representation of the local data based on the parameterization using a generative AI function; evaluating the local update using the synthetic representation to obtain an evaluation result indicative of the performance of the local update; and updating the machine learned model based on the evaluation result and the local update.

Abstract: An object is to provide a video processing system, a video processing apparatus, and a video processing method that can be expected to improve recognition accuracy of an object in a video. A video processing system includes video acquisition means, time difference information acquisition mean, and recognition means. The video acquisition means acquires an input video. The time difference information acquisition means acquires first time difference information between the frames of the input video. The recognition means inputs the input video and the first time difference information between the frames of the input video to a trained recognition model trained using a training video and second time difference information between frames of the training video and recognizing an object in the input video.

Abstract: The decision process of a first machine learning (ML) model may be explained based on a second ML model implemented on an apparatus. The apparatus may obtain a prediction about an image made based on the first ML model. The apparatus may further determine visual concepts associated with the image that may have been used by the first ML model to make the prediction, and determine respective contributions of the visual concepts to the prediction made by the first ML model. The apparatus may then generate, based on the second ML model, a textual description that explains the respective contributions of the visual concepts to the prediction made by the first ML model. The second ML model may determine respective image features associated with the visual concepts, map the determined image features to corresponding text features, and generate the textual description based at least on the text features.

Abstract: An inspection assistance system includes one or more processors including hardware, and the one or more processors are configured to: acquire a video signal including information of a medical image; perform recognition processing based on the video signal, to enable outputting of inspection assistance information for assisting an inspection; acquire identification information included in the video signal, the identification information being different from the information of the medical image; and determine whether the video signal is a video signal for the recognition processing, based on the identification information.

Abstract: The present invention relates to an edge computing system comprising a data processing unit, the data processing unit configured to generate a status data element based on an image of an area, and send the status data element to a remote component, wherein the image comprises an aerial image of the area, wherein the edge computing system, particularly the data processing unit thereof, is configured to receive a data stream from a sensor, and wherein a time interval between receiving the data stream and generating the status data element is less than 30 minutes, preferably less than 10 minutes, further preferably less than 1 minute. The present invention also relates to an associated method, and to an aerial vehicle comprising the edge computing system.

Abstract: A system for faster object attribute and/or intent classification may include an machine-learned (ML) architecture that processes temporal sensor data (e.g., multiple instances of sensor data received at different times) and includes a cache in an intermediate layer of the ML architecture. The ML architecture may be capable of classifying an object's intent to enter a roadway, idling near a roadway, or active crossing of a roadway. The ML architecture may additionally or alternatively classify indicator states, such as indications to turn, stop, or the like. Other attributes and/or intentions are discussed herein.

Abstract: A system includes one or more cameras configured to attach to an aircraft and capture a plurality of images. The plurality of images includes a first image including a runway and a subsequently captured second image including the runway. The system includes an aircraft computing system configured to identify common features in the first and second images, determine changes in locations of the common features between the first and second images, and determine a predicted landing location of the aircraft in the second image based on the changes in locations of the common features. The aircraft computing system is configured to abort landing on the runway based on the predicted landing location relative to the runway.

Abstract: This application provides an image processing method and apparatus, an electronic device, and a computer-readable storage medium. The method includes receiving a prompt that needs to be processed; obtaining a text feature of the prompt, and mapping the text feature to a generability index and a description type of the prompt; in response to the generability index being greater than an index threshold, the description type representing that the prompt does not comprise a verb, and the prompt comprising a plurality of clauses, obtaining similar images respectively corresponding to the plurality of clauses, image-text similarities between the clauses and the corresponding similar images being greater than an image-text similarity threshold; determining an image difference between the similar images respectively corresponding to the plurality of clauses; and in response to the image difference being less than an image difference threshold, using the similar images as illustrations of the corresponding clauses.

Abstract: According to implementations of the subject matter described herein, there is provided a solution for semantic representation of text in a document. In this solution, textual information comprising a sequence of text elements and layout information of the text element are determined from a document. The layout information indicates a spatial arrangement of the plurality of text elements presented within the document. Based at least in part on the plurality of text elements and the layout information, respective semantic feature representations of the plurality of text elements are generated. By jointly using both the textual information and the layout information, rich semantics of the text elements in the document can be effectively captured in the feature representations.

Abstract: A method for performing predictive operations, the method comprising receiving a classification dataset comprising clock drawing images, generating, using a classifier, one or more classification outputs, the one or more classification outputs comprising one or more identifications of dementia or non-dementia for respective ones of clock drawing images. The classifier comprises one or more weights based on a latent space associated with a relevance factor variational autoencoder (RF-VAE). The RF-VAE comprises an encoder configured to generate the latent space. The RF-VAE comprises a decoder configured to generate reconstructions of the second one or more clock drawings based on the latent space. The latent space comprises one or more latent dimensions representative of one or more unique aspects of variation associated with the second one or more clock drawings. The one or more latent dimensions comprise minimal total correlation between the one or more latent dimensions and two dimensions.

Abstract: Computer-implemented methods for real-time analysis of video images are described. In an embodiment, the method includes receiving video images captured by at least one camera; detecting anonymous body regions of individuals in the video images by performing a body key point analysis on the video images using a first neural network; extracting subregions from the video images containing the anonymous body regions; and processing the extracted subregions using a second neural network to classify the anonymous body regions contained in the extracted subregions according to demographic information. Corresponding systems and non-transitory computer-readable media are also described.

Abstract: The present invention relates to a decoder (18) for a race timing system (16), comprising: a reader interface (24) for receiving detections of sports timing transponders (14) passing an antenna (106); a cable connection interface (26) for connecting a network cable and/or a computer interface cable; a user interface (28) for providing configuration feedback to a user; a processing unit (84) for determining passing times of participants (10) in a sports event based on the detections; and a housing (30) for protecting the processing unit from environmental impact, wherein the user interface is arranged on an outside of the housing for providing user access and the cable connection interface is arranged under a protective flap (42) that is configured to provide user access to the cable connection interface in an open state and to shield the cable connection from environmental impact in a closed state. The present invention further relates to a race timing system (16).

Abstract: In an approach to managing telematic data transmission using a global navigation satellite system, one or more computer processors monitor an internet connectivity of a computing device associated with a vehicle. One or more computer processors determine the internet connectivity has failed. One or more computer processors retrieve health data associated with the vehicle. One or more computer processors transform the health data to a format associated with a global navigation satellite system (GNSS). One or more computer processors generate a message in the format associated with the GNSS that includes the transformed health data. One or more computer processors generate a message in the format associated with the GNSS that includes the transformed health data. One or more computer processors transmit the message from the computing device associated with the vehicle to a satellite. One or more computer processors transmit the message from the satellite to a service provider.

Abstract: A fresh produce management system enables customers to acquire single servings of fresh produce while incorporating a food waste prevention app to help inform consumers on how to reduce food waste in their own homes. The storage of the food in the vending machine enhances produce shelf life and maintains food quality. The app itself tells customers when their food is expiring and gives personal recipe recommendations based on consumers' past shopping behaviors.

Abstract: In a locker apparatus, an operation unit includes authentication means, and a control device locks a locking device in order to bring a door into a closed state except when luggage is deposited in a box and when the luggage is taken out from the box. The operation unit includes a reopening door button for giving an instruction to reopen the door. When the reopening door button is operated within 15 minutes after the luggage is deposited in the box, the door is closed, and the locking device is locked once, the control device unlocks the locking device, on a condition that authentication by the authentication means is established, in order to permit the door to be reopened, and locks the locking device again when the door is closed thereafter.

Abstract: According to an aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing a printing control program, the printing control program causing a computer to execute transmission processing, billing processing, and settlement processing. In the transmission processing, print data is transmitted to a printer. In the billing processing, billing is received in a sales server in which a vendor of an operating system of the computer sells an application program. In the settlement processing, settlement is performed when the print data is transmitted and printing is successful in the printer.

Abstract: Game information for a wagering game is received from a remote gaming table. Live actions by a live dealer for the wagering game are captured by an image capture device. Image information based on the captured live actions are transmitted to the remote gaming table to provide a live holographic image of the dealer at the remote gaming table. An input device receives game input from the live dealer and game instructions based on the game input are transmitted to the remote gaming table to execute game actions at the remote gaming table to generate a game result for the wagering game.

Abstract: A gaming system includes an electronic gaming machine (EGM) located within a defined zone of a gaming venue, a transmitter configured to output a first unique identifier, and a server including a processor and a memory storing instructions, where the instructions cause the processor to receive, in connection with a request at a mobile terminal to transfer funds to a gaming wallet accessible at the EGM, a communication from the mobile terminal including the first unique identifier, determine, based on the received first unique identifier, whether the mobile terminal is within the defined zone, and transmit, in response to determining whether the mobile terminal is within the defined zone, a signal to the mobile terminal indicating whether the mobile terminal is within the defined zone.