Abstract: An apparatus includes a setting unit and a control unit. The setting unit performs a first print setting indicating a poster printing function that divides a single piece of image data into a plurality of pieces of image data and generates print jobs, and a second print setting indicating a correction function for correcting calculated gray values when an image processing apparatus converts a plurality of pieces of color image data to grayscale, based on a number of colors acquired from each of the plurality of pieces of color image data. The control unit restricts the setting unit from performing the first print setting and the second print setting. In a state where one of the first print setting and the second print setting is performed by the setting unit, the control unit controls the setting unit not to perform the other print setting.

Abstract: Techniques are described that facilitate training an artificial intelligence (AI) model in a manner that ensures equitable model performance across different sub-groups. According to an embodiment, a system is provided that includes a memory that stores computer executable components, and a processor that executes the computer executable components stored in the memory. The computer executable components include a training component that trains a machine learning (ML) model on training data to perform an inferencing task using an equitable loss function that drives equitable performance of the ML model across different sub-groups represented by the training data, resulting in trained version of the ML model that provides a defined equitable performance level across the different sub-groups. The equitable loss function is “sub-group aware” and penalizes variation in model performance across the sub-groups during model training and validation.

Abstract: Systems and methods for handwriting recognition using language modeling facilitate improved results by using a trained language model to improve results from a handwriting recognition machine learning model. The language model may be a character-based language model trained on a dataset pertinent to field values on which the handwriting recognition model is to be used. A loss prediction module may be trained with the handwriting recognition model and/or the language model and used to determine whether a prediction from the handwriting recognition model should be refined by passing the prediction through the trained language model.

Abstract: Systems and methods for viewing, storing, transmitting, searching, and editing application-specific audiovisual content (or other unstructured data) are disclosed in which edge devices generate content on the fly from a partial set of instructions rather than merely accessing the content in its final or near-final form. An image processing architecture may include a generative model that may be a deep learning model. The generative model may include a latent space comprising a plurality of latent codes and a trained generator mapping. The trained generator mapping may convert points in the latent space to uncompressed data points, which in the case of audiovisual content may be generated image frames. The generative model may be capable of closely approximating (up to noise or perceptual error) most or all potential data points in the relevant compression application, which in the case of audiovisual content may be source images.

Abstract: Methods, apparatuses, electronic devices, and mediums related to classification models are provided. The method may include: obtaining a set of positive samples and a set of negative samples, a sample in the set of positive samples including a pair of images from a first image set annotated as similar with each other, and a sample in the set of negative samples including a pair of images annotated as dissimilar with each other from the first image set; training a first sub-network based on the set of positive samples and the set of negative samples; constructing a classification model including a second sub-network, where initial weights of the second sub-network are weights of corresponding nodes of the trained first sub-network; and training the classification model using a second image set that includes at least one image that is common with at least one image in the first image set.

Abstract: An implement management system detects implement wear and monitors implement states to modify operating modes of a vehicle. The system can determine implement wear using the pull of the implement on the vehicle, the force and angle of which is represented by an orientation vector. The system may measure a current orientation vector and determine an expected orientation vector using sensors and a model (e.g., a machine learned model). Additionally, the implement management system can determine an implement state based on images of the soil and the implement captured by a camera onboard the vehicle during operation. The system may apply different models to the images to determine a likely state of the implement. The difference between the expected and current orientation vectors or the determined implement state may be used to determine whether and how the vehicle's operating mode should be modified.

Abstract: A non-transitory recording medium storing a computer program for an application providing a Graphical User Interface (GUI) causing a computer to perform each step of a control method of the computer, the method including: a reception step of receiving a print ticket including a print setting according to activation of the program for the application by an operating system after a print instruction is given by a user; and a display control step of displaying a screen for authenticated printing based on the received print ticket in which the authenticated printing has been set.

Abstract: A method for determining a stereotactic brain target includes at least one target point, the method comprising the following steps: selecting patients for whom the result measured following treatment at at least one target point is greater than or equal to a threshold, postoperative imaging having been performed for each of the patients; processing the postoperative imaging so as to determine coordinates of the at least one target point; selecting brain marker points; processing the postoperative imaging so as to determine coordinates of the marker points; creating a learning database comprising the coordinates of the target points and the coordinates of the marker points determined for all of the selected patients; determining a prediction function giving the coordinates of at least one target point according to the coordinates of the marker points by using the learning database and a supervised statistical learning method; processing preoperative imaging of a new patient to be treated so as to determine coo

Abstract: In some example embodiments, a computer-implemented method may include training a machine learning model in a first database instance using a machine learning algorithm and a training dataset in response to receiving a request to train, serializing the trained machine learning model into a binary file in response to the training of the machine learning model, recreating the trained machine learning model in a second database instance using the binary file in response to receiving a request to apply the machine learning model, and generating an inference result by applying the recreated trained machine learning model on the inference dataset in the second database instance.

Abstract: Techniques are provided for computing systems, methods and computer program products to produce efficient image-to-image translation by adapting unpaired datasets for supervised learning. A first model (a powerful model) may be defined and conditioned using unsupervised learning to produce a synthetic paired dataset from the unpaired dataset, translating images from a first domain to a second domain and images from the second domain to the first domain. The synthetic data generated is useful as ground truths in supervised learning. The first model may be conditioned to overfit the unpaired dataset to enhance the quality of the paired dataset (e.g. the synthetic data generated). A run-time model such as for a target device is trained using the synthetic paired dataset and supervised learning. The run-time model is small and fast to meet the processing resources of the target device (e.g. a personal user device such as a smart phone, tablet, etc.

Abstract: A computing device may receive a set of user documents. Data may be extracted from the documents to generate a first graph data structure with one or more initial graphs containing key-value pairs. A model may be trained on the first graph data structure to classify the pairs. Until a set of evaluation metrics for the model exceeds a set of deployment thresholds: generating, a set of evaluation metrics may be generated for the model. The set of evaluation metrics may be compared to the set of deployment thresholds. In response to a determination that the set of evaluation metrics are below the set of deployment thresholds: one or more new graphs may be generated from the one or more initial graphs in the first graph data structure to produce a second graph data structure. The first and second graph can be used to train the model.

Abstract: A method includes obtaining, using at least one processor, audio/video content. The method also includes processing, using the at least one processor, the audio/video content with a trained attention-based machine learning model to classify the audio/video content. Processing the audio/video content includes, using the trained attention-based machine learning model, generating a global representation of the audio/video content based on the audio/video content, generating a local representation of the audio/video content based on different portions of the audio/video content, and combining the global representation of the audio/video content and the local representation of the audio/video content to generate an output representation of the audio/video content. The audio/video content is classified based on the output representation.

Abstract: A data sharing system for sharing datasets of data providers to data consumers and transferring incentives from the data consumers to the data providers in response to the data-sharing. The system includes a multi-angle alliance guided data valuation module for fair allocation of the incentives between the data consumers. The system also includes a flexible-scenario routed dataset comparison module for evaluating the data provided by the data providers via one of a plurality of evaluating routes. The system provides enhanced use of computer cloud and enables both data alliance and growing capacity of artificial intelligence (AI) supermodels for sustainable data sharing. Moreover, the system uses coreset based Shapley valuation method for efficient data valuation.

Abstract: An apparatus for controlling object tracking and a method therefor are provided. The apparatus includes an object detector configured to detect an object in an image, an object tracker configured to track the object, a learning device configured to learn whether to enable the object detector based on features of the object and tracking results of the object tracker, and a controller configured to determine whether to enable the object detector by interworking with the learning device.

Abstract: A method in one embodiment includes receiving, at an edge device, new data for training a model, the edge device having stored distilled data used to represent historical data to train the model, the historical data being stored in a remote device, and the amount of the historical data being greater than the amount of the distilled data. The method further includes training the model based on the new data and the distilled data. With the data processing solution of this embodiment, the model can be trained at the edge device with fewer storage resources based on the distilled data, thereby achieving higher model accuracy.

Abstract: An oral cavity scanning device is provided in the invention. The oral cavity scanning device includes an image capturing unit, an IMU circuit and a processing unit. The image capturing unit obtains a first image and a second image. The IMU circuit obtains IMU information corresponding to the first image and the second image. The processing unit obtains a distance value between the first image and the second image. The processing unit uses a contour algorithm to obtain a first contour and a second contour. The processing unit obtains first sampling points according to the first contour and second sampling points according to the second contour. The processing unit uses a feature algorithm to find relative feature points between the first sampling points and the second sampling points. The processing unit uses a depth information algorithm to obtain the depth information of each feature point.

Abstract: Examples are provided for traffic sign detection systems. In one example, a traffic detection system in a vehicle includes an image sensor, a communication system, a processor, and a storage device storing instructions executable by the processor to capture an image of an environment of the vehicle via the image sensor, process the image to detect visual information regarding a traffic sign in the image, the visual information indicating a recognized sign, receive cryptographic data via the communication system from a transmitter associated with the traffic sign, the cryptographic data including a cryptographic representation of a traffic sign signal for controlling the traffic sign, and selectively control one or more vehicle systems of the vehicle based on a cryptographic verification of the recognized sign using the cryptographic data.

Abstract: Disclosed is an air screen detector device. The air screen detector device includes a display for displaying an air screen image in one direction by using visible light, an infrared light source disposed on the display to provide infrared light in the same direction as that of the visible light, an infrared image sensor disposed on one side of the display and the infrared light source to receive the infrared light reflected by a finger provided in the air screen image disposed on the other side of the infrared light source, a first multifocal lens disposed between the infrared image sensor and the infrared light source to provide the infrared light to the infrared image sensor, and a visible light filter disposed between the first multifocal lens and the infrared image sensor to remove the visible light.

Abstract: An image processing method includes: extracting, based on pixel values of a plurality of pixels of an image obtained, a plurality of first current pixels having pixel values included in a first color range including a specific color range that defines a specific color, from the plurality of pixels; excluding, in second region different from a first region including a center region of the image, (i) a low saturation pixel from the plurality of first current pixels when the number of low saturation pixels is greater than or equal to a first threshold value, and (ii) a high saturation pixel, which has a pixel value included in a high saturation range, from the plurality of first current pixels when the number of high saturation pixels is greater than or equal to a second threshold value; and outputting a plurality of second current pixels obtained as a result of the excluding.

Abstract: A control method for a scanning system that reduces load on an information processing apparatus when transmitting scanned data to a cloud server requiring an access token is provided. The control method includes acquiring, by an information processing apparatus, an access token from an authorization server, transmitting, by the information processing apparatus, the access token and an address of a cloud server in a first transmission, receiving, by an image forming apparatus, the address and the access token transmitted in the first transmission, scanning, by the image forming apparatus, a document based on the reception of the address and the access token, and transmitting, by the image forming apparatus, the image data generated by scanning the document and the access token to the cloud server in a second transmission.