Nbcuniversal Media, LLC

Michael Shane Wilkinson

Robert Glenn Deen

Braxton Paul Perkins

Anthony Christopher Anderson

Catherine Mary Graichen

Corey Nicholas Bufi

Zhaohui Harry Sun

Budhaditya Deb

Anthony James Hoogs

Roderic Greene Collins

Thomas Bause

Aaron Shaw Markham

Sheau Bao Ng

Jenny Marie Weisenberg

Eric J Avila

Borzou Afshar

Andrew Lucas

Jenny Marie Weisenberg Williams

A tangible, non-transitory machine-readable medium includes machine readable instructions that, when executed by processing circuitry, cause the processing circuitry to receive an inquiry regarding digital content. The inquiry is indicative of a request for the digital content or a request for a determination to be made regarding the digital content. The instructions, when executed, also cause the processing circuitry to process the inquiry by analyzing aggregated result data generated using a series of neural network layers based on the digital content and metadata associated with the digital content. Furthermore, the instructions, when executed, cause the processing circuitry to generate response data to the inquiry using a neural network based on analyzing the aggregated result data.

Artificial intelligence-assisted content source identification

Systems and methods for controlling transmission of digital content stored at a server are disclosed. According to at least one embodiment, a method includes: receiving, at the server, session information from a device, the session information including information for authorizing the server to decrypt the digital content for transmission to a playback device; determining, at the server, to decrypt the digital content for the transmission to the playback device, based on the session information; decrypting, at the server, the digital content for the transmission to the playback device, in response to determining to decrypt the digital content; and transmitting, at the server, the decrypted digital content to the playback device.

Digital rights management for cinema systems

Systems and methods for optimal variable bit rate packing are disclosed. Some embodiments include encoding media at a plurality of bit rates to generate a corresponding plurality of encodings, where each of the plurality of encodings include a plurality of framesets, generating advance encoding data for each of the plurality of encodings, where each advance encoding data includes information related to a data size of the plurality of framesets of the corresponding encoding, transmitting the generated advance encoding data to a client, receiving a request from the client including a request for one or more framesets of a first encoding, and transmitting the requested one or more framesets of the first encoding to the client.

System and method for optimal variable bit rate packing

Digital content, such as video, audio, multimedia, and similar files may be segmented at a determined frame resolution rate and analyzed to determine a fingerprint for each content segment. An identifier that uniquely identifies each content segment may be determined based on the fingerprint of the respective content segment. The identifier may be associated with data relating to the content segment and with one or more identifiers that uniquely identify other content segments of the content. The content segment identifiers may be used to construct a content segment sequencing representation that represents the playback of the content segments in playback order.

Fingerprint-defined segment-based content delivery

A method and system for identifying a source of a copied work that in one embodiment includes obtaining at least some portions of a reference work, collecting at least some portions of the suspect work, matching the suspect work with the reference work, wherein the matching includes temporally aligning one or more frames of the reference work and the suspect work, spatially aligning frames of the reference work and the suspect work, and detecting forensic marks in the suspect work by spatiotemporal matching with the reference work.

System and method for forensic analysis of media works

Digital content, such as video, audio, multimedia, and similar files may be associated with metadata that is in some manner descriptive of the content by means of a unique identifier generated based on the content or a portion of the content. The two may be packaged in a wrapper. If the content is separated from the metadata, such as by circulation in a non-compliant environment, the content may be recognized by a regeneration of the identifier. Based upon the re-created identifier, the metadata may be re-associated with the content. Other scenarios may include generation of compilations for files that are not included in a wrapper with content and metadata, content that becomes dissociated with metadata by transcoding, and so forth.

Metadata/content re-association system and method

A computer implemented method for generating a receipt. The method includes, accessing a universal digital fingerprint associated with an item of content, wherein the fingerprint is invariant across one or more formats of the item of content. The method includes accessing an identification value by the processor. The method includes generating a digital token by cryptographically binding the digital fingerprint and the identification value, wherein the digital token is invariant across the one or more formats of the item of content. The method further includes associating the digital token with at least one right to the item of content.

Method and system for validating rights to digital content using a digital token

A system and method are provided for analyzing 3D digital content to determine whether a watermark is detectable. The watermark may exist in 2D content that is converted to 3D, and in such cases, the survivability of the watermark to the conversion process is evaluated. An anticipated location of the watermark in left and right 3D images may be determined, and the detectability based upon the anticipated location. A report may indicate whether the watermark survived the conversion in one or both images, or neither. The process may be performed for single frames, sequences of single frames, or entire files containing many image frames. Watermark placement may also be proposed for locations in 2D content, 3D content, or both. Watermarks may similarly be placed in the content.

Growing community of inventors

Newhall, CA, United States of America

Michael Shane Wilkinson