International Business Machines Corporation

New York University

Kyndryl, Inc.

Suman Sedai

Rahil Garnavi

Dwarikanath Mahapatra

Pallab K Roy

Bhavna Josephine Antony

Kerry J Halupka

Jorge Andres Moros Ortiz

Hiroshi Ishikawa

Rajib Chakravorty

Seyedbehzad Bozorgtabar

Stefan Von Cavallar

Xi Liang

Stefan Renard Maetschke

Bruno De Assis Marques

Ruwan B Tennakoon

Hsin-Hao Yu

Timothy Michael Lynar

John Michael Wagner

Stephen Michael Moore

Hidemasa Muta

Khoi-Nguyen Dao Tran

Kimiko Wilson

Zongyuan Ge

Noel Garry Faux

Turgay Tyler Kay

Sree Harish Maathu

In an approach to improve multi-factor authentication embodiments generate an evaluation-mask over one or more modified items on a modified image created by a generative adversarial network (GAN). Further, embodiments create a scoring grid by comparing an original image with the modified image to identify different pixels between the original image and the modified image, and overlay the evaluation-mask over the identified different pixels on the modified image. Embodiments display the modified image as a multi-factor authentication prompt to a user and prompt the user to provide a response that identifies one or more modifications in the modified image. Additionally, embodiments compute an evaluation score based on a comparison of the response from the user with the evaluation-mask, to validate the response from the user, and authenticate and grant the user access to data or other resources if the evaluation score meets or exceeds a predetermined threshold.

Generating an evaluation-mask for multi-factor authentication

A generative adversarial network including a generator portion and a discriminator portion is constructed. The network is configured such that the network operates to enhance intensity images, wherein an intensity image is obtained by illuminating an object with an energy pulse and measuring the return strength of the energy pulse, and wherein a pixel of the intensity image corresponds to the return strength. As a part of the configuring, a loss function of the generative adversarial network is minimized, the loss function comprising a mean square error loss measurement of a noisy intensity image relative to a mean square error loss measurement of a corresponding clean intensity image. An enhanced intensity image is generated by applying the minimized loss function of the network to an original intensity image, the applying improving an image quality measurement of the enhanced intensity image relative to the original intensity image.

Neural network based enhancement of intensity images

In an approach to generating conversational image representations, one or more computer processors detect one or more utterances by a user, wherein utterances are either textual or acoustic. The one or more computer processors generate one or more image representations of the one or more detected utterances utilizing a generative adversarial network restricted by one or more user privacy parameters, wherein the generative adversarial network is fed with an extracted sentiment, a generated avatar, an identified topic, an extracted location, and one or more user preferences. The one or more computer processors display the generated one or more image representations on one or more devices associated with respective one or more recipients of the one or more utterances.

Autocreation of conversational image representation

In an aspect for generating device-specific OCT image, one or more processors may be configured for receiving, at a unified domain generator, first image data corresponding to OCT image scans captured by one or more OCT devices; processing, by the unified domain generator, the first image data to generate second image data corresponding to a unified representation of the OCT image scans; determining by a unified discriminator, third image data corresponding to a quality subset of the unified representation of the OCT image scans having a base resolution satisfying a first condition and a base noise type satisfying a second condition; and processing, using a conditional generator, the third image data to generate fourth image data corresponding to device-specific OCT image scans having a device-specific resolution satisfying a third condition and a device-specific noise type satisfying a fourth condition.

Normalizing OCT image data

Content of entertainment media that is being consumed by a user is analyzed. An element of the content that is of a first character is identified. A preference associated with the user to consume entertainment media that contains elements of a second character is identified. An updated version of the element is generated. The updated version of the element is of the second character, such that the media is consumed by the user with the element in the updated version.

Content modification based on element contextualization

Aspects of the invention include a computer implemented method for simulating visual field test results from structural scans, the method includes processing eye image data to extract visual functioning related features. Additionally, generating a representation of a visual function of the eye that is independent of a visual field test (VFT) configuration. Then generating a simulated VFT configuration specific test result based at least in part on the representation.

Simulating visual field test from structural scans

Aspects of the invention include systems and methods that train a teacher neural network using labeled images to obtain a trained teacher neural network, each pixel of each of the labeled images being assigned a label that indicates one of a set of classifications. A method includes providing a set of unlabeled images to the trained teacher neural network to generate a set of soft-labeled images, each pixel of each of the soft-labeled images being assigned a soft label that indicates one of the set of classifications and an uncertainty value associated with the soft label, and training a student neural network with a subset of the labeled images and the set of soft-labeled images to obtain a trained student neural network. Student-labeled images are obtained from unlabeled images using the trained student neural network.

Uncertainty guided semi-supervised neural network training for image classification

A retinal structure and function forecasting method, system, and computer program product include producing an enriched feature representation of clinical measurements and clinical data combined with optical coherence tomography (OCT) data, training a forecasting model with the enriched feature representation, and forecasting a retinal structure at a forecast date based on the trained forecasting model.

Early detection and management of eye diseases by forecasting changes in retinal structures and visual function

Aspects of the invention include obtaining data regarding a plurality of devices in a viewing environment and analyzing a content item to be displayed in the viewing environment. Aspects also include identifying an interaction between a scene of the content item and at least one of the plurality of devices based at least in part upon the analyzing and identifying a viewer in the viewing environment and obtaining a user profile for the viewer. Based upon the interaction and the user profile, aspect include activating the at least one of the plurality of devices during playback of the scene. Aspects further include monitoring one or more characteristics of the viewer during playback of the scene and updating the user profile for the viewer based on the one or more characteristics.

Identifying and calibrating internet of things (IOT) devices to enhance content consumption experience

A method, computer system, and computer program product for real-time pediatric eye health monitoring and assessment are provided. The embodiment may include receiving a plurality of real-time data related to an individual's eye health from a user device. The embodiment may also include assessing biometric indications relating to eye health based on the plurality of real-time data. The embodiment may further include generating a report on the assessed biometric indications. The embodiment may also include collecting clinical information from one or more databases. The embodiment may further include determining whether the assessed biometric indications reach pre-configured threshold conditions. The embodiment may also include generating alerts and recommendations based on analysis of the collected clinical information and the assessed biometric indications based on the assessed biometric indications satisfying the pre-configured threshold conditions.

Real-time passive monitoring and assessment of pediatric eye health

An apparatus for automatically separating and collecting a fluid stream into multiple portions, including a tubing to receive the fluid stream, a plurality of valves attached to the tubing in spaced relation along the length of the tubing and a plurality fluid collection containers attached to the valves. A controller controls the activation of the plurality of valves to collect and separate portions of the fluid stream in the plurality of the fluid collection containers. A flow meter detects the flow of the fluid stream in the tubing and measures a flow rate of the fluid stream. The controller activates the valves based on the detection of the fluid and the flow rate. The controller also activates the valves based on the timing of the fluid stream. The controller also monitors the flow rate over time and activates the valves in sequence based on the flow rate.

Automatic control of the separation and collection of fluid

From a first image using a model, a first uncertainty map is generated. An uncertainty level of a location in the first uncertainty map corresponds to a detection of a known structure in a portion of the first image. A first weighted image corresponding to the first uncertainty map is generated, the generating including assigning a first weight to a pixel of the first image, the first weight corresponding to the uncertainty level of a location in the first uncertainty map corresponding to the pixel. From a second image using a model, a second uncertainty map is generated. A second weighted image corresponding to the second uncertainty map is generated. The first image and the second image are combined to form a composite image, each image participating in the composite image according to the corresponding weighted image.

Uncertainty region based image enhancement

A system for registering and segmenting images includes an image scanner configured to acquire an image pair including a first image at a first time and a second image at a second time that is after the first time. A joint registration and segmentation server receives the image pair from the image scanner and simultaneously performs joint registration and segmentation on the image pair using a single deep learning framework. A computer vision processor receives an output of the joint registration and segmentation server and characterizes how a condition has progressed from the first time to the second time therefrom. A user terminal presents the characterization to a user.

Growing community of inventors

Melbourne, Australia

Suman Sedai