U.S. Bank National Association

Varshini Sriram

Soumitri Naga Kolavennu

Priyanka Singhal

Lavanya Basavaraju

Jayant Chawla

Rodolfo Carriedo Roque

Bryan Lee Stuhlsatz

Ankur Tomar

Aspects of the embodiments described herein are related to systems, methods, and computer products for performing computer-aided detection of fraud device installation events, especially at an Automated Teller Machine (ATM). Aspects of embodiments described herein provide artificial intelligence systems and methods that detect the presence of obstructions and persons to determine when a fraud device installation event occurs. The fraud detection system performs object detection and can determines whether a detected access event has an associated transaction to determine that a fraud device installation event occurs. The fraud detection system can also track the time the camera view is obstructed, the activity time of the person standing at the monitored device when no transaction occurs, and detect objects that resemble fraud devices to determine if a fraud device installation event occurs.

Systems and methods for video-based fraud detection

Systems and methods are provided for access control enforcement through real-time packet inspection and transaction screening. The system employs one or more processors and memory to execute instructions that analyze packet headers and payloads, enforce access control policies, and utilize machine learning-driven agentic systems to predict associations with sanctioned entities. By comparing packet data against sanction lists and employing hierarchical decision-making processes, the system blocks or flags high-risk packets.

System and method for performing access control enforcement via packet inspection

Aspects of the embodiments described herein are related to systems, methods, and computer products for performing video redaction. When performing video redaction, a video is received and converted into a plurality of frames. For each of the frames, it is detected if one or more people are present in at least one of the plurality of frames. Bounding regions are determined for the detected people in each frame. The bounding regions are tagged with an identifier identifying the person associated with each bounding region. An icon identifying the detected person is determined and displayed on an interface as a selectable input, wherein each selectable input is selectable to redact or keep the detected person in the video. Once a selection to redact a person is received, the bounding regions of the selected person are filled with replacement pixels. The plurality of frames are then converted into a new video.

Video redaction

Geometric extraction is performed on an unstructured document by recognizing textual blocks on at least a portion of a page of the unstructured document, generating bounding boxes that surround and correspond to the textual blocks, determining search paths having coordinates of two endpoints and connecting at least two bounding boxes, and generating a graph representation of the at least a portion of the page, the graph representation including the plurality of textual blocks, the coordinates of the vertices of each bounding box and the coordinates of the two endpoints of each search path.

Growing community of inventors

Menlo Park, CA, United States of America

Varshini Sriram