AT&T Intellectual Property I, LP

At&t Mobility II LLC

Liveclips LLC

At&t Mobolity II LLC

Eric Petajan

Subhabrata Bapi Sen

Emir Halepovic

David Eugene Weite

Douglas W Vunic

Shichang Xu

Eric Martin Zavesky

James H Pratt

Manuel Briand

Ting-Hsiang Tony Hwang

Jing Sasha Guo

Jessica R Owensby

Sreemanananth Sadanand

Xavier Riley

Kunwar Handa

Kevin Culwell

Arlene Cava

Niklas Carlsson

Vengatanathan Krishnamoorthi

Esteban Gomez

Lucian Jiang-Wei

Zachary Jared Bell

Nicholas Covey

Rajakumar Adhikari

Facilitating avatar modifications for learning and other videotelephony sessions in advanced networks is provided herein. Operations of a system include evaluating a recorded interaction associated with a first entity during consumption of a first portion of a video conference determined to include the first entity. The operations also can include transforming an actual representation of the first entity in the recorded interaction to an avatar representation, resulting in an edited interaction of the first entity. Further, the operations can include outputting the edited interaction of the first entity for consumption of a second portion of the video conference by rendering the edited interaction for a second entity.

Facilitating avatar modifications for learning and other videotelephony sessions in advanced networks

A processing system may obtain a downscaled version of a reference copy of a video, comprising a plurality of downscaled versions of a plurality of frames of the reference copy of the video, obtain a first recorded frame of a first variant of a plurality of variants associated with the reference copy of the video, where the plurality of variants comprises a plurality of copies of the video encoded at different bitrates, generate a first downscaled version of the first recorded frame, calculate a first plurality of image distances between the first downscaled version of the first recorded frame and the plurality of downscaled versions of the plurality of frames of the reference copy of the video, and determine a first frame index of the first recorded frame in accordance with a first least image distance from among the first plurality of image distances that is calculated.

Adaptive bitrate video testing from screen recording

Personalized content delivery (e.g., using a computerized tool) is enabled. For example, a system can comprise: a processor and a memory that stores executable instructions that, when executed by the processor, facilitate performance of operations, comprising: determining a quality of service metric representative of a quality of service of streamed content delivered via a content delivery network, determining a quality of experience metric representative of a quality of experience associated with a user profile of a consumer of the streamed content, and based on the quality of service and the quality of experience, modifying the streamed content.

Content delivery system

Aspects of the subject disclosure may include, for example, a method that includes obtaining, by a processing system including a processor, video frames over a network; the processing system uses a machine learning algorithm to identify in each frame a first region comprising a natural image and a second region comprising a synthetic graphic image. The processing system separates the natural image from the synthetic graphic image to generate a natural video and a graphics video, encodes the natural video, and processes the graphics video to generate instructions for rendering graphic images at a client system. The client system performs a decoding procedure for the encoded video, a rendering procedure for client-side graphics in accordance with the instructions, and a compositing procedure to obtain a presentable video stream including the natural image and a client-side graphic corresponding to the synthetic graphic image. Other embodiments are disclosed.

Separation of graphics from natural video in streaming video content

Aspects of the subject disclosure may include, for example, a method in which a processing system acquires a video image and voice data of a remote viewer of a live event content, and generating animation parameters relating to the video image. An avatar of the viewer is constructed based on the animation parameters; the avatar includes an animated version of the video image. The animated version and voice data are encoded to obtain a compressed animation data stream and a compressed audio stream, which are transmitted at a low bit rate to a remote system providing the content. The remote system aggregates compressed animation data streams and compressed audio streams to obtain a virtual audience for the event; the virtual audience includes avatars of the remote viewers and sound produced by the remote viewers. Other embodiments are disclosed.

Virtual audience using low bitrate avatars and laughter detection

Concepts and technologies disclosed herein are directed to audio and video synchronization. According to one aspect disclosed herein, an audio-video ('AV') synchronization system can simultaneously capture samples of a pre-encode media stream and a post-encode media stream. The pre-encode media stream can include AV content prior to being encoded. The post-encode media stream can include the AV content after being encoded. The AV synchronization system can align a pre-encode video component of the pre-encode media stream with a post-encode video component and can determine a video offset therebetween. The AV synchronization system can align a pre-encode audio component of the pre-encode media stream with a post-encode audio component of the post-encode media stream and can determine an audio offset therebetween. The AV synchronization system can then compare the video offset and the audio offset to determine if the post-encode media stream is synchronized with the pre-encode media stream.

Audio and video synchronization

A buffer classification system calculates first flow metrics for a first flow in response to receiving first packet level data. The system in response to receiving HTTP information, generates buffer emulation data and creates one or more data training sets using the first flow metrics and buffer emulation data, wherein training data used to create the one or more training data sets is further generated by computing an average throughput per second observed over different time windows during a video playback. The system in response to creating the one or more training data sets, generates one or more classifier rules used to determine a buffer condition of a flow. The system in response to receiving second packet level data, calculates second flow metrics for a second flow and predicts a buffer condition for the second flow based on the second flow metrics and the one or more classifier rules.

System for determining a predicted buffer condition based on flow metrics and classifier rules generated in response to the creation of training data sets

A processing system may obtain a first frame of a first variant associated with a reference copy of a video, where a plurality of variants comprises copies of the video encoded at different bitrates, determine a frame number of the first frame from a visual overlay containing the first frame number, calculate a first image distance between the first frame and a frame of the reference copy having the frame number, determine, from among a plurality of image distances for frames of each variant having the frame number, a closest image distance to the first image distance, the plurality of image distances comprising image distances between frames of the variants and the frame of the reference copy having the frame number, and identify the first frame as being from the first variant in accordance with a variant associated with the first closest image distance.

Adaptive bitrate video testing

Concepts and technologies are disclosed herein for adaptive bit rate mobile video objective testing. A system can receive a plurality of screen-capture frames, where each of the plurality of screen-capture frames corresponds to a respective frame of an adaptive bit rate video stream being displayed on a screen of a test device. The system can create a screen-capture video stream and determine that the screen-capture video stream corresponds to a reference video stream that is non-annotated. The system can obtain a reference video signature package, align the screen-capture video stream with the reference video stream using the reference video signature package, generate full reference video quality performance indicators, and determine delivery quality performance indicators for the screen-capture video stream. The system can join the full reference video quality performance indicators and the delivery quality performance indicators to form an objective quality of experience data structure.

Adaptive bit rate mobile video objective testing

A system for annotating frames in a media streamincludes a pattern recognition system (PRS)to generate PRS output metadata for a frame; an archivefor storing ground truth metadata (GTM); a device to merge the GTM and PRS output metadata and thereby generate proposed annotation data (PAD); and a user interfacefor use by the human annotator HA. The user interfaceincludes an editorand an input deviceused by the HAto approve GTM for the frame. An optimization systemreceives the approved GTM and metadata output by the PRS, and adjusts input parameters for the PRS to minimize a distance metric corresponding to a difference between the GTM and PRS output metadata.

Annotating media content for automatic content understanding

A method to correct for temporal variability in incoming streams of media and data to optimize the performance of a pattern recognition system includes the steps of receiving from one of the incoming streams a point in time when an event is announced, applying probability distribution about the point in time, shifting a point of highest probability of the probability distribution back in time by an amount effective to accommodate for a delay between the event and the announcement, comparing a detected pattern of the event to a stored pattern of similar events and applying a confidence value to the comparison, and confirming to the pattern recognition system that the event occurred at the point of highest probability when the confidence score exceeds a predefined threshold. The method is useful to determine the time at which a particular play occurs during a sporting event, such as the time of a shot-on-goal in a soccer match.

Growing community of inventors

Watchung, NJ, United States of America

Eric Petajan