Abstract: A system includes an electronic display, a computer processor, one or more memory units, and a module stored in the one or more memory units. The module is configured to access a source image stored in the one or more memory units and determine depth data for each pixel of a plurality of pixels of the source image. The module is further configured to map, using the plurality of pixels and the determined depth data for each of the plurality of pixels, the source image to a four-dimensional light field. The module is further configured to send instructions to the electronic display to display the mapped four-dimensional light field.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a neural network to predict optical flow. One of the methods includes obtaining a batch of one or more training image pairs; for each of the pairs: processing the first training image and the second training image using the neural network to generate a final optical flow estimate; generating a cropped final optical flow estimate from the final optical flow estimate; and training the neural network using the cropped optical flow estimate.

Abstract: Disclosed is an automatic image analysis method that can be used to automatically recognise at least one rare characteristic in an image to be analysed. The method comprises a learning phase during which at least one convolutional neural network is trained to recognise characteristics, a parameter space of dimension n, in which n?2, is constructed from at least one intermediate layer of the network, a presence probability function is determined for each characteristic in the parameter space from a projection of reference images in the parameter space. During a phase of analysing the image to be analysed, the method comprises a step of recognising the at least one rare characteristic in the image to be analysed on the basis of the presence probability function determined for the at least one rare characteristic.

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed. An example apparatus includes an audio detector to determine a first audience count based on signatures of audio data captured in the media environment, a thermal image detector to determine a heat blob count based on a frame of thermal image data captured in the media environment, and an audience image detector to identify at least one audience member based on a comparison of a frame of audience image data with a library of reference audience images, the audience image detector to perform the comparison in response to the first audience count not matching the heat blob count.

Abstract: An information processing device includes: a storage unit that stores first information indicating at least one of a structure or a physical property of a first object obtained by capturing a real physical body into a virtual space; and a specifying unit that specifies an arrangement of a second object indicating a virtual object in the virtual space so that the arrangement is capable of expressing an interaction with the first object, based on an arrangement condition of the second object and on the first information.

Abstract: A system for identifying handwritten characters on an image using a classification model that employs a neural network. The system includes a computer having a processor and a memory device that stores data and executable code that, when executed, causes the processor to read and convert typed text on the image to machine encoded text to identify locations of the typed text on the image; identify a location on the image that includes handwritten text based on the location of predetermined typed text on the image; identify clusters of non-white pixels in the image at the location having the handwritten text; generate an individual and separate cluster image for each identified cluster; classify each cluster image using machine learning and at least one neural network to determine the likelihood that the cluster is a certain character; and determine the accuracy of the characters by comparing to a secondary database.

Abstract: An object of the present invention is to acquire biometric information on small animals without being affected by the movements of the small animals and also without giving stress to the small animals. The present invention includes the steps of: acquiring pieces of image information on a hand and/or a foot of a quadrupedal small animal; extracting a predetermined piece of signal information from among the pieces of image information that have been acquired; and processing the predetermined piece of the signal information which has been extracted so as to acquire biometric information on the small animal. The present invention further includes the step of capturing an image of an abdominal side of the small animal, and the step of capturing an image of the small animal from below a location of the small animal.

Abstract: This invention provides a division pattern determining apparatus for determining a division pattern for dividing a region to be encoded in image data using a neural network, comprising: a determining portion configured to execute on the image data of the region to be encoded a determination for a plurality of blocks of different sizes whether or not to divide each block using a neural network and determine a division pattern of one or more blocks included in the region to be encoded on the basis of the determination, wherein the neural network of the determining portion is constituted by a plurality of layers; and wherein when a size of a block for which whether or not to divide is to be determined is large, information from an intermediate layer positioned in an upper position is used.

Abstract: A method for retrieving heights of densely-covered forest canopies near power transmission lines includes: acquiring ICESat-2 LiDAR data, JL-1 image data, auxiliary data and three-dimensional information data about the power transmission lines within a target area; carrying out image preprocessing on a JL-1 image to generate a first image; screening the ICESat-2 LiDAR data according to a screening rule to obtain high-quality laser tree height data; by employing the first image, the high-quality laser tree height data and the auxiliary data, training a neural network model for retrieving a forest tree height according to an optical image and the auxiliary data; by employing the neural network model, generating a height distribution map of densely-covered trees in the target area; and calculating a height difference between a tree and a power transmission line nearest the tree to generate a hidden danger troubleshooting theme map.

Abstract: Current pretrained vision-language models for cross-modal retrieval tasks in English depend upon on the availability of many annotated image-caption datasets for pretraining to have English text. However, the texts are not necessarily in English. Although machine translation (MT) tools may be used to translate text to English, the performance largely relies on MT's quality and may suffer from high latency problems in real-world applications. Embodiments herein address these problems by learning cross-lingual cross-modal representations for matching images and their relevant captions in multiple languages. Embodiments seamlessly combine cross-lingual pretraining objectives and cross-modal pretraining objectives in a unified framework to learn image and text in a joint embedding space from available English image-caption data, monolingual corpus, and parallel corpus. Embodiments are shown to achieve state-of-the-art performance in retrieval tasks on multimodal multilingual image caption datasets.

Abstract: A method for automatically identifying faulty signatures for autonomous driving applications, the method includes receiving, by a processing circuit, a signature; matching the signature to a group of first signatures that are untagged and are randomly obtained; identifying, based on the matching, first top matching signatures; matching the signature to a group of second signatures that are untagged and are correctly or erroneously indicative of a detection of reference elements; identifying, based on the matching of the signature to the second group of second signatures, second top matching signatures; determining an overlap between the first top matching signatures and the second top matching signatures; and determining whether the signature is faulty or faultless based on the overlap.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a IT machine learning model. One of the methods includes receiving training data comprising a plurality of unlabeled training inputs and a plurality of labeled training inputs; generating augmented training data, comprising generating, for each of the plurality of unlabeled training inputs, a respective augmented training input by applying a data augmentation technique to the unlabeled training input; and training the machine learning model on the augmented training data. In particular, but not exclusively, the model may be trained for perceptual tasks (e.g. tasks relating to vision or speech).

Abstract: A multiple dwelling house interphone system includes a collective entrance machine, a dwelling room master device, and a management master device. The collective entrance machine is for calling a visit destination by a visitor. The dwelling room master device is for responding to the call by a dweller. The management master device is for speaking with the dweller by a manager. The management master device includes a speech permissibility storage section and a speech control section. The speech permissibility storage section registers a permissibility of calling and speaking to the dwelling room master device for each dwelling unit. The speech control section controls the calling and speaking. The speech control section does not perform the calling control when a calling operation by the management master device is performed towards the dwelling room master device of a dwelling unit registered as “unpermitted” in the speech permissibility storage section.

Abstract: Devices, systems, and methods are provided for brand matching using multi-layer machine learning. A method may include generating, based on a first embedding vector and a second embedding vector as inputs to a twin neural network, a third embedding vector and a fourth embedding vector; generating, based on the first embedding vector and the second embedding vector as inputs to a difference neural network, a difference vector indicative of a difference between the first embedding vector and the second embedding vector; generating a concatenated vector by concatenating the third embedding vector with the fourth embedding vector and the difference vector; generating, based on the concatenated vector as an input to a feedforward neural network (FFN), a score between zero and one, the score indicative of a relationship between a first entity and a second entity.

Abstract: Disclosed are an astronomical telescope stand, an auxiliary calibration method for the astronomical telescope and an astronomical telescope system. The astronomical telescope includes at least two rotating shafts perpendicular to each other, each rotating shaft is provided with an encoder; an electronic camera shooting in the same direction as the telescope to be mounted; a display screen; a processor that connects with the encoder, the electronic camera and the display screen; the processor includes an image information parsing module, an electronic image generation module and a coordinate correction module. The telescope is mounted on the stand, and the electronic camera shoots the same image as the image observed by the telescope. The image information parsing module solves the image to obtain the center coordinate and the field size of the image. The electronic image generation module generates the simulated image according to the center coordinate and the field size.

Abstract: An information processing apparatus in the present invention includes: an acquisition unit that, based on registration information including a plurality of registrants and registration time associated with each of the plurality of registrants and on the current time, acquires a first biometrics information group including biometrics information on a candidate of a matching process from a registered biometrics information group including biometrics information on the plurality of registrants; and a matching unit that matches biometrics information on a person detected from an image captured in a matching area against biometrics information included in the first biometrics information group.

Abstract: An image processing apparatus 1 includes a pre-processor 10 and a main processor 30. The pre-processor 10 generates pre-processing data by performing a pre-processing of Bayer data outputted from an imaging device 91, and outputs the generated pre-processing data. The main processor 30 performs an image processing based on the pre-processing data.

Abstract: A computer-implemented method includes receiving, by a computing device, a particular textual description of a scene. The method also includes applying a neural network for text-to-image generation to generate an output image rendition of the scene, the neural network having been trained to cause two image renditions associated with a same textual description to attract each other and two image renditions associated with different textual descriptions to repel each other based on mutual information between a plurality of corresponding pairs, wherein the plurality of corresponding pairs comprise an image-to-image pair and a text-to-image pair. The method further includes predicting the output image rendition of the scene.

Abstract: An event detection system to operations that include: detecting an event based on at least a portion of the sensor data, the event comprising a plurality of event attributes; determining one or more event attributes from the plurality of event attributes of the event transgress a threshold value; and causing display of a notification that includes a presentation of at least the portion of the sensor data in response to the determining that the one or more event attributes from among the plurality of event attributes of the event transgress the threshold value.

Abstract: In image processing system comprises an input configured to receive an image signal, the image signal including a plurality of frames of image data; and a processor configured to automatically determine an image classification based on at least one frame of the plurality of frames, and dynamically generate a mapping metadata based on the image classification. The processor includes determination circuitry configured to determine a content type for the image signal; segmentation circuitry configured to segment the image data into a plurality of feature item regions, based on the content type; extraction circuitry configured to extract at least one image aspect value for respective ones of the plurality of feature item regions.