Motorola Corporation

Arris Enterprises LLC

Google Inc.

Motorola Mobility LLC

General Instrument Corporation

Arris Technology, Inc.

Google Technology Holdings LLC

Bhavan Gandhi

Faisal Ishtiaq

Raghavan Subramaniyan

Kevin Lynn Baum

Stephen P Emeott

Anthony J Braskich

Zhu Li

Andrew J Aftelak

Isselmou Ould Dellahy

Benedito J Fonseca, Jr

Kevin J O'Connell

Dongge Li

Michael L Needham

Renxiang Li

Aggelos K Katsaggelos

Brett L Lindsley

Alfonso Martinez Smith

Crysta J Metcalf

Sanjeev Kumar Mishra

Sridhar Kunisetty

Varma L Chanderraju

Vinay Kalra

Bharath R Rao

David D Nicozisin

Benedito J Fonseca

Raghavan A104 Admiralty Ave Apts Subramaniyan

Methods and systems are provided for bitrate adaptation of a video asset to be streamed to a client device for playback. The method includes selecting a representation from a manifest which expresses a set of representations available for each chunk of the video asset and generating a dynamic manifest for the video asset in which the representation selected for the at least one chunk is recommended for streaming to the client device. The selection of the representation recommended for the chunk may be based on at least one of historic viewing behavior of previous viewers of the chunk, content analysis information for the chunk, a level of available network bandwidth, a level of available network storage, and data rate utilization information of network resources including current, average, peak, and minimum data rate of network resources.

Method for dynamically managing content delivery

Systems and methods for generating alerts and enhanced viewing experience features using on-screen data are disclosed. Textual data corresponding to on-screen text is determined from the visual content of video data. The textual data is associated with corresponding regions and frames of the video data in which the corresponding on-screen text was detected. Users can select regions in the frames of the visual content to monitor for a particular triggering item (e.g., a triggering word, name, or phrase). During play back of the video data, the textual data associated with the selected regions in the frames can be monitored for the triggering item. When the triggering item is detected in the textual data, an alert can be generated. Alternatively, the textual data for the selected region can be extracted to compile supplemental information that can be rendered over the playback of the video data or over other video data.

Generating alerts based upon detector outputs

A system and method are provided for recommending a segment of a segmented video asset of particular interest to a client. A copy of a video asset is created such that the copy is in the form of a set of segments for being transmitted to a client device for playback. A relationship is established between start and end times of each segment relative to a standard version of segments of the video asset, and metadata is generated for each segment of the copy. The metadata and relationship is used relative to the standard version with viewing data collected across a population of viewers having viewed the video asset to produce viewing metrics for each segment of the set of segments of the copy. The viewing metrics are provided to a recommender which uses the viewing metrics to generate a recommendation of a segment of the copy to a client.

Recommendation of segmented content

A method receives video content and metadata associated with video content. The method then extracts features of the video content based on the metadata. Portions of the visual, audio, and textual features are fused into composite features that include multiple features from the visual, audio, and textual features. A set of video segments of the video content is identified based on the composite features of the video content. Also, the segments may be identified based on a user query.

Content based video content segmentation

Continuity of an entire user session (including the primary content stream, secondary content streams, and user context) is preserved so that the user can resume the session at a later time, at a different place, and, possibly, using different equipment. When a user pauses a session, the context of that session is automatically preserved. Upon resumption, the session begins where the user left off, resuming the primary media stream at the point where the user stopped, knowing what secondary content items the user has already seen, and re-establishing any user-set parameters for the session (e.g., playback volume, allocation of streams to particular screen real estate, whether closed captioning is turned on, and the like). For time-shifted content consumption, the system intelligently selects, filters, and processes contextual information (such as characteristics of the primary media) in order to present companion streams that are relevant and engaging to the user.

Preserving a consumption context for a user session

Disclosed is a method of operating a secondary device in a manner associated with operation of a primary device including obtaining first information corresponding to a media asset being output by the primary device, processing the first information to determine local media-signature information, transmitting the first information for receipt by a server, receiving secondary information from the server, wherein the secondary information includes a plurality of asset-media signatures that respectively correspond to respective portions of the media asset, attempting to determine a time-based correlation between at least one portion of the local media-signature information and at least one of the asset-media signatures, and outputting one or more portions of time-relevant asset streams from the secondary device, the one or more portions being determined at least indirectly based upon the correlation.

Time-synchronizing a parallel feed of secondary content with primary media content

Disclosed are methods and apparatus for providing information to a first client device (e.g., a tablet computer) for presentation on that device. The information may be related to multimedia content (e.g., a television program) that may be presented using a second client device (e.g., a television). Firstly, an activity level for a portion of the multimedia content is determined. Using the activity level, an amount of the information is assigned to that portion of the multimedia content. The amount of the information assigned is dependent on that determined activity level. The assigned information is then provided for use by (e.g., for display on) the first client device.

Information provision

An object detection algorithm that generates a two-layer Gaussian Mixture Model (GMM) during a training session, and subsequent to the training session, uses the two-layer GMM to perform face detection. No labeling of local features is needed. The only input that is provided by a user is the setting of a few global parameters for the image being captured during the training session, such as, for example, the person's facial pose.

Method and apparatus for performing object detection

A method and apparatus for performing a predictive motion search in a video encoder system using motion vectors representing the difference in coordinates of a macroblock of data in a current frame of video data and coordinates of a related macroblock of data in a reference frame of video data. A plurality of motion vector predictors is obtained where the motion vector predictors represent approximations of possible motion vectors for a current macroblock. A search pattern is defined. Each motion vector predictor of the plurality of motion vector predictors is searched around using the search pattern. A final motion vector is then determined.

Method and apparatus for performing high quality fast predictive motion search

A processor () utilizes information regarding one or more physical dimensions of an individual () to better inform a personal identification process. In one embodiment, the measured physical dimensions are utilized to influence the conduct of a face recognition process. In one embodiment, a Bayesian Belief Network can be utilized to facilitate such processes.

Personal identification method and apparatus

A search system () for a database () including records having a multiple disparate types of media is provided. The search system supports queries, that include different types of search criteria, including content based retrieval search criteria. A fuzzy logic method () is provided for effectively combining the results of different types of search criteria. The fuzzy logic method also allows confidence levels entered by the user for search criteria to be considered in combining results. Retrieval relevance values for documents for at least some search criteria are used in the fuzzy logic method. For content based image retrieval searches, the retrieval relevance values are computed by mapping a distance between quantitative characterizations of a search basis image, and other images into a finite range.

Method for fuzzy logic rule based multimedia information retrival with text and perceptual features

A method and apparatus for determining the quality of a block match for a candidate motion vector in a video encoder system using motion vectors representing the difference in coordinates of a macroblock of data in a current frame of video data and coordinates of a related macroblock of data in a reference frame of video data. The method can include defining a search pattern, searching a region based on the search pattern for a candidate motion vector for evaluation, calculating a difference metric, calculating a bias based on the difference between a predictor motion vector and a candidate motion vector, determining a modified difference metric by adding the difference metric to the bias, and defining a final motion vector based on the modified difference metric.

Method and apparatus for determining block match quality

At least one visual significance metric is determined () for at least some frames belonging to an original series of frames (). Key frames are identified () as a function, at least in part, of the visual significance metric. Cumulative visual significance values are then determined () for at least some of the frames that intervene between each pair of key frames. These cumulative visual significance values are then used to identify () frames of additional content interest. Various frames are then selected () for use in a visual summary. In one embodiment, all of the key frames and frames of additional content interest are selected for inclusion in the visual summary.

Growing community of inventors

Vernon Hills, IL, United States of America

Bhavan Gandhi