Abstract: One embodiment disclosed relates to a system for detecting anomalous messaging, discovering compromised accounts, and generating responses to threatened attacks. The system utilizes API commands and log forwarding for interaction and communication between a messaging and account hunting platform, other hunting platforms, an action center, and a security operations center. Another embodiment relates to a method of, and system for, performing a complete root cause analysis. Another embodiment relates to a method of, and system for, anomaly discovery which may advantageously utilize reference data to correlate different anomalies for reporting as a single incident.

Abstract: An email attempting to perpetrate a business email compromise (BEC) attack is detected based on similarity of the email to a known BEC email and on similarity of the email to a user email that would have been sent by the purported sender of the email. Metadata of the email is extracted and input to a BEC machine learning model to find the known BEC email among BEC email samples. The extracted metadata are also input to a personal user machine learning model of the purported sender to generate the user email.

Abstract: Unwanted web contents are detected in an endpoint computer. The endpoint computer receives a web page from a website. The reputation of the website is determined and the web page is scanned for malicious codes to protect the endpoint computer from web threats. To further protect the endpoint computer from web threats including mutating unwanted web contents, page structure traits of the web page are generated and compared to page structure traits of other web pages detected to contain unwanted web contents.

Abstract: A method of tracking an inventory of objects via a mobile communications device includes acquiring an image of one or more of the objects via the mobile communications device, which also collects a location of the mobile communications device while acquiring the image of the one or more of the objects. The location and image are transferred from the mobile communications device to a remote server via a wireless network, such that the one or more of the objects are identified at the server based on the image, and the location and identity of the one or more objects are stored on a database associated with the server.

Abstract: A method of transmitting a live video stream and a pre-recorded video stream over a wireless network to a mobile device, the method including: receiving the live video stream and the pre-recorded video stream for a current scheduling window; dividing the live video stream and the pre-recorded video stream into segments in the current scheduling window, each segment including an aggregate data amount, a transmission start time and a transmission end time; determining a transmission schedule for the current scheduling window, said transmission schedule including the segments; locating slack time slots, the slack time slots being time slots for which no segment is allocated; and scheduling bursts associated with the pre-recorded video stream in the slack time slots.

Abstract: A method for encoding video for streaming includes receiving a plurality of sequential image frames generated by a 3D graphics rendering engine. Graphics rendering contexts are obtained, including pixel depth map, rendering camera parameters, and camera motion from the 3D rendering engine. The method next entails selecting key frames among the plurality of sequential image frames, interpolating non-key frames via 3D image warping, and encoding all key frames and warping residues of non-key frames. The system is implementable on a server linked to a mobile user device for receiving the encoded frame data. The mobile user device is configured to decode the encoded frame data and display a corresponding image to a user of the mobile user device.

Abstract: The described system and method provide joint rate control and scalable stream adaptation for multiple clients concurrently competing for the same access networks. For each such client, an optimization problem is constructed and solved to determine the streaming rate over each access network, the video packets to be transmitted, and the access network over which each video packet is sent. The rate control and stream adaptation problem is constructed as an integer program in an embodiment of the invention, with an objective to minimize a cost function of the expected video distortion. Randomized packet scheduling is accounted for in an embodiment of the invention by relaxing the integer program into real-valued optimization programs and deriving convex programming approximations.

Abstract: A method of transmitting a live video stream and a pre-recorded video stream over a wireless network to a mobile device, the method including: receiving the live video stream and the pre-recorded video stream for a current scheduling window; dividing the live video stream and the pre-recorded video stream into segments in the current scheduling window, each segment including an aggregate data amount, a transmission start time and a transmission end time; determining a transmission schedule for the current scheduling window, said transmission schedule including the segments; locating slack time slots, the slack time slots being time slots for which no segment is allocated; and scheduling bursts associated with the pre-recorded video stream in the slack time slots.

Abstract: A method for encoding video for streaming includes receiving a plurality of sequential image frames generated by a 3D graphics rendering engine. Graphics rendering contexts are obtained, including pixel depth map, rendering camera parameters, and camera motion from the 3D rendering engine. The method next entails selecting key frames among the plurality of sequential image frames, interpolating non-key frames via 3D image warping, and encoding all key frames and warping residues of non-key frames. The system is implementable on a server linked to a mobile user device for receiving the encoded frame data. The mobile user device is configured to decode the encoded frame data and display a corresponding image to a user of the mobile user device.

Abstract: A method of tracking an inventory of objects via a mobile communications device includes acquiring an image of one or more of the objects via the mobile communications device, which also collects a location of the mobile communications device while acquiring the image of the one or more of the objects. The location and image are transferred from the mobile communications device to a remote server via a wireless network, such that the one or more of the objects are identified at the server based on the image, and the location and identity of the one or more objects are stored on a database associated with the server.

Abstract: The described system and method provide joint rate control and scalable stream adaptation for multiple clients concurrently competing for the same access networks. For each such client, an optimization problem is constructed and solved to determine the streaming rate over each access network, the video packets to be transmitted, and the access network over which each video packet is sent. The rate control and stream adaptation problem is constructed as an integer program in an embodiment of the invention, with an objective to minimize a cost function of the expected video distortion. Randomized packet scheduling is accounted for in an embodiment of the invention by relaxing the integer program into real-valued optimization programs and deriving convex programming approximations.

Abstract: A method for detecting indicators of misleading content in a markup language coded document is provided. The method includes extracting a set of tags from the markup language coded document. The method also includes combining tags to create a tag structure signature. The tag structure signature is configured to include a set of n-grams. Each of the set of n-grams includes at least two tags from the set of tags. The method further includes comparing the tag structure signature against a set of known bad tag structure signatures to determine similarity.

Abstract: In one embodiment, a fingerprint is generated for each document (e.g., e-mail, web page) containing markup language (e.g., HTML) code. The fingerprint is indicative of the structure of the markup language code in the document. The fingerprint may be formed by extracting markup language tags from the document and then linking together the extracted tags to form a single string. The fingerprint may be hashed through a hashing function to generate a signature key that may be used to create a directory for the document and other documents having the same fingerprint. The grouping of documents with the same fingerprint facilitates creation of anti-spam rules or identification of web pages from particular websites, for example.