Bank of America Corporation

Sundarrajan Raman

Jeff L Howard

Padmanabham Ponnada

Igor Yago

Oscar Lopez-Martinez

Siva Chaitanya Thurlapati Venkata

Ramamurthy Pavan Nallamothu

Prasad Dasari

Vinu Viswanathan

Sivasundaram Umapathy

Dharmendra Golkonda

Manikandan Selvaraj

John Popplewell

Ram Sekar

Vikas Shriniwas Shahapure

Prasanth Kaliyangad

Ankita Jain

Nagalakshmi Tirumalaikumar

Venkatesh Sivaraman

A synthetic media detection system stores, for each of a set of predefined objects an object repository with previously determined characteristics of predefined objects and an action description repository with natural language descriptions of actions that are known to be possible to occur with the predefined object. The system receives at least a portion of a media stream including a video to be presented on a media player device. An object is detected in the received portion of the media stream and properties of the detected object are determined. An action is determined that is associated with the detected object. An object probability is determined that the detected object is a synthetic media object. A chronological description is determined of the actions associated with the detected object. An overall probability that the object is synthetic media object using the chronological description is determined and displayed with the video.

Action oriented detection of synthetic media objects

A firewall system stores filtering criteria which include rules for blocking presentation of all or a portion of the media content based at least in part on an identity of an individual appearing in the media content. The firewall system receives the media content. The firewall system determines the identity of the individual appearing in the media content. Based at least in part on the identity of the individual appearing in the media content and the filtering criteria, the firewall system determines an action for securing the media content. The action may be allowing presentation of the media content, blocking presentation of the media content, or blocking presentation of a portion of the media content. The determined action is automatically implemented.

Image-based firewall for synthetic media prevention

A computing device generates a call in a programming language of a computing application requesting a feature of videos stored in a media repository. A query system receive the call and determines a command associated with obtaining the feature requested by the call. The determined command corresponds to an image analysis to perform on at least a portion of the stored videos. The query system determines, based at least in part on the determined command, an artificial intelligence model to execute on at least the portion of the stored videos. The query system determine, by executing the determined artificial intelligence model, a model output that includes the requested feature. The query system provides, in the programming language of the computing application, an indication of the requested feature.

System for performing programmatic operations using an image-based query language

A system receives an audio-video recording that includes image frames. For each image frame, a portion of the image frame is identified that corresponds to a first face of a first subject. An image region is determined for the identified portion of the image frame. A first face portion of the audio-video recording is stored that corresponds to the determined image region for each of the image frames along with a private key for the audio-video recording and a first mask identifier for the first subject. A public key is generated for the private key. A remaining portion of the audio-video recording is associated with the generated public key and the first mask identifier for the first subject. The remaining portion, the public key, and the first mask identifier are provided to a media server that communicates with a media player.

Media hosting system employing a secured video stream

A secure media hosting system stores a protected portion of an audio-video (AV) recording, a private key associated with the AV recording, and a mask identifier. A media player device is communicatively coupled to the secure media hosting system. The media player device receives a media stream that includes a first portion of an AV recording, a public key corresponding to the AV recording, and the mask identifier. The public key and mask identifier are extracted from the media stream. A request for the protected portion of the AV recording is provided. The request includes the extracted public key and mask identifier. After providing the request for the protected portion of the AV recording, the protected portion of the AV recording is received. The first portion and protected portion of the AV recording are combined. The combined first and protected portions of the AV recording are displayed.

Video player for secured video stream

A system includes a media source and a firewall system. The media source stores media content that includes one or both of an image or a video. The firewall system stores filtering criteria that include rules for blocking communication of all or a portion of the media content based at least in part on an identity of an individual appearing in the media content. The firewall system receives a request for the media content. The firewall system determines characteristics of a user that provided the request for the media content. The firewall system determines the identity of the individual appearing in the media content. Based at least in part on the characteristics of the user that provided the request for the media content, the identity of the individual appearing in the media content, and the filtering criteria, the firewall system determines an action for securing the media content.

Image-based firewall system

A resiliency controller may monitor execution of a plurality of services, each service of the plurality of services associated with a different container of a containerized computing system. In response to requests received from a plurality of connected client computing systems, the resiliency controller predicts, based on historical container operation information, a first predicted latency associated with the first request and a second predicted latency associated with the second request. The resiliency controller associates a first latency response to the first request and a second latency response to the second request, and initiates a selected latency response based on an indication that at least one of the first predicted latency or the second predicted latency has been exceeded.

Resiliency control engine for network service mesh systems

A media source is configured to provide a media stream that includes an audio-video recording. The audio-video recording comprises image frames. A video presentation device receives the media stream provided by the media source and selects a portion of the image frames of the media stream. The device determines, based at least in part on characteristics of the video presentation device and properties of the selected portion of the image frames, a processing resource for executing an algorithm configured to determine a probability that the media stream includes synthetic media. The selected portion of the frames and an indication of the algorithm are provided to the processing resource. The processing resource executes the algorithm to determine the probability that the media stream includes synthetic media. The audio-video recording and a representation of the probability that the media stream includes synthetic media are displayed.

Media player and video verification system

A media source is configured to provide a media stream that includes an audio-video recording. A first algorithm detects a presence of synthetic media in a video based on a first feature type. A second algorithm detects the presence of synthetic media based on a second feature type. A device determines that the first feature type is presented in a selected portion of the image frames but the second feature type is not. In response, the first algorithm is executed using the selected portion of the image frames as an input. Based on the executed first algorithm, a probability is determined that the media stream provided by the media source includes synthetic media.

Automated video verification

A user device transmit a natural language query with a request for a description of videos stored in a media repository. A query system receives the query and determines a command associated with obtaining the requested description of the videos stored in the media repository requested by the query. The determined command corresponds to an image analysis to perform on at least a portion of the stored videos. The query system determines, based at least in part on the determined command, an artificial intelligence model to execute on at least the portion of the stored videos. The query system determines, by executing the determined artificial intelligence model, a model output that includes the requested description of the videos stored in the media repository. The query system provides a response to the query. The response includes the requested description of the videos stored in the media repository.

Image-based query language system for performing database operations on images and videos

A system includes an execution engine and a processor. The execution engine receives application code that, when executed, is configured to generate a second data element from a first data element that is stored in a first database and store the second element in a second database. The execution engine converts the code into an execution plan and executes it. The execution plan includes a first operation to obtain the first element from the first database, a second operation to apply a transformation to the first element, and a third operation to store the first element as the second element in the second database. The processor accesses the execution plan and determines that the third operation includes storing the second element. In response, the processor generates a data lineage for the second element by extracting, from each operation from the third to the first, a portion of the data lineage.

System and method for automatically obtaining data lineage in real time

Systems for dynamically transforming data are provided. Database data may be received and ingested into a system. Ingesting the data may include executing one or more first data governance functions, such as data quality evaluation functions, data controls, and the like. The ingested data may then be output for further processing as first processed data and first data governance information may be captured and stored. The first processed data may be processed to execute one or more data transformations. Data transformations may include calculations, formatting, derivations, and the like. In some arrangements, second data governance functions may be executed on the transformed data. The transformed data may then be output as second processed data. The system may capture second data governance information as the data is transformed. The second processed data may then be published to one or more downstream databases for use in one or more applications executed by an entity.

Dynamic data transformation system

Aspects of the disclosure relate to self-correcting dependent software upgrades. A computing platform may scan a software source code and parse release notes associated with the source code. The computing platform may identify a code modification to be performed on dependent source code, and may identify a code segment of the dependent source code. Then, the computing platform may retrieve, based on a machine learning model, a code fix for the code segment, and a test case associated with the code modification. The computing platform may update the code segment by applying the code fix, and may merge the updated code segment with the dependent source code. Then, the computing platform may run the test case on the modified source code. Then, the computing platform may determine an outcome for the modified source code, and may initiate, based on the outcome, an action related to the dependent source code.

Growing community of inventors

Tamil Nadu, India

Sundarrajan Raman