Abstract: A computer implemented method, system, and non-transitory computer-readable device that may be used in a remote deposit environment. Upon receiving a user request, based on interactions with the UI, the method implements an electronic deposit of a financial instrument by activating a camera on the client device to generate a live video stream of image data of a field of view of at least one camera, wherein the live video stream includes imagery of at least a portion of each side of the financial instrument. The method continues by extracting data fields based on the formation of image objects of each side of the financial instrument from the live video stream of image data. A flip detection pauses the forming of image objects during the flip. The extracted data fields are communicated to a remote deposit server to complete the remote deposit.

Abstract: Normalized resolutions are determined for first and second regions of interest of an initial video stream captured by a video capture device located within a physical space. The first region of interest is associated with a first conference participant within the physical space and the second region of interest is associated with a second conference participant within the physical space. Instructions are transmitted to the video capture device to cause the video capture device to capture, at the normalized resolutions, a first video stream associated with the first region of interest and a second video stream associated with the second region of interest. The first and second video streams conform sizes and quality levels of the first and second conference participants within separate user interface tiles of a conferencing software user interface to which the first and second video streams are output.

Abstract: A system including at least one aerially mounted camera and at least one processor performs a method for selecting a location to place a virtual tripwire onto grid points of a geocoordinate grid. The processor(s) directs a camera to capture a series of images within the camera's field of view. The processor(s) identifies one or more objects of interest within the images and determines movements of the objects through the series of images. The processor(s) identifies directions of travel of the objects of interest. The processor(s) further determines a vector of direction of travel for each object of interest on the grid. The processor(s) then determines at least one candidate location for placement of the virtual tripwire and automatically selects a location for placement of the virtual tripwire based at least on the vector of direction of travel for an object of interest.

Abstract: A technology for remote object imaging processing. An example of the technology can include calculating geographic coordinates for an object using in part a line-of-sight distance to the object from a camera coupled to a vehicle, calculating an angle of the sun relative to the object using sun position data that corresponds to a time and date and the geographic coordinates for the object, determining a white balance setting based in part on the angle of the sun relative to the object, and applying the white balance setting as a parameter used to process an image of the object captured using the camera coupled to the vehicle.

Abstract: An eyeball detection device includes an imaging device that obtains a face image which includes the left and right eyeball images of the target person; a first distance calculating unit that calculates a first-type distance representing the distance between the target person and the imaging device; and an eyeball searching unit that performs a search for the left and right eyeball images from the face image and, when only one of the left and right eyeball images of the target person is detected, searches for the other eyeball image of the target person based on the search result regarding the already-detected eyeball image as obtained by the eyeball detecting unit, based on the first-type distance, and based on a second-type distance representing the distance between the left and right eyeballs of the target person.

Abstract: A system configured to communicate with a plurality of premises monitoring systems. The system includes at least one processor configured to, for each premises monitoring system of a first subset of the premises monitoring systems, receive premises data comprising flood sensor data indicating that a potential water leak, receive user confirmation data confirming that a water leak event occurred at the premises, identify at least one camera associated with the premises monitoring system that captured video of at least a portion of the water leak event, and collect the video captured by the at least one camera. A machine learning (ML) model is trained using the video collected from each premises monitoring system of the first subset of the premises monitoring systems. The ML model is deployed to at least a second subset of the plurality of premises monitoring systems for detecting a water leak.

Abstract: A method for detecting data defects and a computing device applying the method obtains a test image for analysis. A field to which the test image relates is determined. Based on the field, a target convolutional layer is determined from a convolutional neural network. The target convolutional layer is used to extract features of the test image. A target score of the test image and a score threshold corresponding to the field are determined. If the target score is less than the score threshold, it is determined that the test image reveals defects, thereby improving an accuracy of defect detection.

Abstract: There is provided an information processing apparatus including a projection control unit (150) that causes visual information to be projected onto a projection surface. The visual information corresponds to content that is provided to the projection surface and disposed on a first object. The content is identified on the basis of a recognition result of the first object and a recognition result of a second object. The second object is provided along with the first object.

Abstract: A camera apparatus includes control circuitry configured to acquire an input color image of an agricultural field, smoothen the input color image with a median blur and convert the smoothened input color image into a plurality of different color spaces. The control circuitry is configured to execute a set of channel operations on an individual channel or combined channels in each color space of the plurality of different color spaces and generate a normalized image based on outputs received from each color space processing path. The control circuitry is configured to determine a threshold value based on a histogram of the normalized image and apply the determined threshold value to generate a first binary mask image. The control circuitry is configured to apply one or more morphology operations to remove noise in the first binary mask image and generate an output binary mask image of foliage mask.

Abstract: A method for viewing pathology images in a web browser is provided.

Abstract: Systems and methods for depth estimation in accordance with embodiments of the invention are illustrated. One embodiment includes a method for estimating depth from images. The method includes steps for receiving a plurality of source images captured from a plurality of different viewpoints using a processing system configured by an image processing application, generating a target image from a target viewpoint that is different to the viewpoints of the plurality of source images based upon a set of generative model parameters using the processing system configured by the image processing application, and identifying depth information of at least one output image based on the predicted target image using the processing system configured by the image processing application.

Abstract: Implementations generally recognize clinical activity using multiple cameras. In some implementations, a method includes obtaining a plurality of videos of a plurality of objects in an environment. The method further includes determining one or more key points for each object of the plurality of objects. The method further includes recognizing activity information based on the one or more key points. The method further includes computing workflow information based on the activity information.

Abstract: One or more highlights of a video stream may be identified. The highlights may be segments of a video stream, such as a broadcast of a sporting event, that are of particular interest to one or more users. According to one method, at least a portion of the video stream may be stored. The portion of the video stream may be compared with templates of a template database to identify the one or more highlights. Each highlight may be a subset of the video stream that is deemed likely to match the one or more templates. The highlights, an identifier that identifies each of the highlights within the video stream, and/or metadata pertaining particularly to the one or more highlights may be stored to facilitate playback of the highlights for the users.

Abstract: The technology disclosed relates to a system and method of exporting learned features between federated endpoints whose learning is confined to respective training datasets. The system includes logic to access a first training dataset to train a first federated endpoint and a second training dataset to train a second federated endpoint. The first and second training datasets have first and second sample sets that share one or more shared sample features. The shared sample features are common between the first and second sample sets. The system includes logic to train a first generator on the first federated endpoint. The system includes logic to use the first trained generator for a second inference on a second performance task executed on the second federated endpoint.

Abstract: According to one embodiment, an information processing device according to an embodiment includes an acquisition unit, a detection unit, and a calculation unit. The acquisition unit acquires image data in which an image of a space where a plurality of specific objects exist is captured. The detection unit detects a position and an orientation of each of the plurality of specific objects included in the image data in the space. The calculation unit calculates, for any evaluation target object of the plurality of specific objects, a close-contact evaluation value indicating a degree of close contact between the evaluation target object and one or a plurality of other specific objects other than the evaluation target object among the plurality of specific objects, based on the position and the orientation of each of the plurality of specific objects.

Abstract: A text recognition system receives a prompt and, based on the prompt, causes a trained region encoder to determine a first region of interest of an image file. The system modifies a first image associated with the first region of interest (e.g., parsed out from the first region) to generate a data augmentation entity that includes a modified image. Using a trained instance encoder, the system generates a first set of visual instances corresponding to the first region of interest image and a second set of visual instances corresponding to the data augmentation entity. The system generates the corresponding first and second sequences. By executing a self-supervised contrastive loss function on the first and second sequences, the system automatically updates a continual knowledge distillation model of the trained region encoder. The system provides the first sequence to an instance decoder to generate output text in response to the prompt.

Abstract: An aspect of the present disclosure provides an apparatus for extracting a vascular function including an information reception unit configured to extract an original CT image from brain-related information received from the outside; an NIFTI image transformation unit configured to transform the original CT image into an NIFTI file format image to acquire time sequence data; a time interpolation unit configured to apply time interpolation to the original CT image through the time sequence data to transform the original CT image into each time-specific 3D CT image; a vessel segmentation unit configured to predict a vessel segmentation mask by passing the each time-specific 3D CT image through a deep learning-based vessel segmentation deep-learning model 141 and generate a 4D vessel mask image by stacking the 3D CT images based on a time axis; and a vascular function extraction unit configured to extract a vascular function from a vessel region of the 4D vessel mask image and calculate a blood flow parameter us

Abstract: A portable variable focus microscope includes a housing having a lower section and an upper section secured together. A lens is removably disposed on at least one lens retention feature of the upper section. The scope further includes a slide platform and a height adjuster, where rotation of the height adjuster within the scope housing adjusts a relative distance between the slide platform and the lens to adjust a focus of the lens. The scope may be used in conjunction with a mobile data device to capture an image of a sample on a slide received in the scope. A program including a CNN algorithm may be employed to identify contaminants in the sample.

Abstract: Provided according to an embodiment of the present invention is a device for managing the environment of a breeding farm, the device comprising: an imaging unit for capturing a thermal image of the inside of a breeding farm; an object detection unit for detecting an object from the captured image; an area division unit for dividing the captured image into multiple areas; a filtering unit for extracting, from the divided areas, a second area except a first area where the object is detected; and a control unit for comparing the temperature of the second area with a reference temperature to detect an abnormal situation.

Abstract: Exemplary embodiments relate to the automatic generation of captions for visual media, including photos, photo albums, non-live video, and live video. The visual media may be analyzed to determine contextual information (such as location information, people and objects in the video, time, etc.). A system may integrate this information with information from the user's social network and a personalized language model built using public-facing language from the user. The personalized language model captures the user's way of speaking to make the generated captions more detailed and personalized. The language model may account for the context in which the video was generated. The captions maybe used to simplify and encourage content generation, and may also be used to index visual media, rank the media, and recommend the media to users likely to engage with the media.