Matthew Grant Berry

US Patent:

8281231, Oct 2, 2012

Filed:

Sep 13, 2010

Appl. No.:

12/881026

Inventors:

Matthew G. Berry - Raleigh NC, US
Changwen Yang - Morrisville NC, US

Assignee:

Digitalsmiths, Inc. - Durham NC

International Classification:

G06F 17/00

US Classification:

715201

Abstract:

Method, systems, and computer program products for synchronizing text with audio in a multimedia file, wherein the multimedia file is defined by a timeline having a start point and end point and respective points in time therebetween, wherein an N-gram analysis is used to compare each word of a closed-captioned text associated with the multimedia file with words generated by an automated speech recognition (ASR) analysis of the audio of the multimedia file to create an accurate, time-based metadata file in which each closed-captioned word is associated with a respective point on the timeline corresponding to the same point in time on the timeline in which the word is actually spoken in the audio and occurs within the video.

US Patent:

8311344, Nov 13, 2012

Filed:

Feb 17, 2009

Appl. No.:

12/372561

Inventors:

Heather Dunlop - Raleigh NC, US
Matthew G. Berry - Raleigh NC, US

Assignee:

Digitalsmiths, Inc. - Durham NC

International Classification:

G06K 9/62

US Classification:

382224

Abstract:

The present disclosure relates to systems and methods for classifying videos based on video content. For a given video file including a plurality of frames, a subset of frames is extracted for processing. Frames that are too dark, blurry, or otherwise poor classification candidates are discarded from the subset. Generally, material classification scores that describe type of material content likely included in each frame are calculated for the remaining frames in the subset. The material classification scores are used to generate material arrangement vectors that represent the spatial arrangement of material content in each frame. The material arrangement vectors are subsequently classified to generate a scene classification score vector for each frame. The scene classification results are averaged (or otherwise processed) across all frames in the subset to associate the video file with one or more predefined scene categories related to overall types of scene content of the video file.

US Patent:

8311390, Nov 13, 2012

Filed:

May 14, 2009

Appl. No.:

12/466167

Inventors:

Matthew G. Berry - Raleigh NC, US

Assignee:

Digitalsmiths, Inc. - Durham NC

International Classification:

H04N 9/82

US Classification:

386249, 386248, 386250

Abstract:

The present disclosure relates to systems and methods for identifying advertisement breaks in digital video files. Generally, an advertisement break identification module receives a digital video file and generates an edge response for each of one or more frames extracted from the video file. If one of the generated edge responses for a particular frame is less than a predefined threshold, then the module identifies the particular frame as the start of an advertisement break. The module then generates further edge responses for frames subsequent to the identified particular frame. Once an edge response is generated for a particular subsequent frame that is greater than the threshold, it is identified as the end of the advertisement break. The video file may then be manipulated or transformed, such as by associating metadata with the advertisement break for a variety of uses, removing the advertisement break from the video file, etc. Optionally, various time and/or frame thresholds, as well as an audio verification process, are used to validate the identified advertisement break.

US Patent:

8380045, Feb 19, 2013

Filed:

Oct 9, 2008

Appl. No.:

12/248596

Inventors:

Matthew G. Berry - Raleigh NC, US
Schuyler E. Eckstrom - Beaufort SC, US

International Classification:

H04N 9/80
H04N 5/76
H04N 5/225
H04N 5/917
H04N 7/173
H04N 11/02
G06F 15/16
G06F 17/00
G06K 9/36
G06K 9/64

US Classification:

386248, 386257, 386329, 3482071, 3482201, 3482311, 37524001, 37524012, 37524027, 382236, 382278, 706 45, 709231, 715234, 725112, 725113

Abstract:

Systems and methods are provided for generating unique signatures for digital video files to locate video sequences within a video file comprising calculating a frame signature for each frame of a first video; and for a second video: calculating a frame signature for each frame of the second video for corresponding first video frame signatures, calculating a frame distance between each of the corresponding video frame signatures, determining video signature similarity between the videos, and searching within a video signature similarity curve to determine a maximum corresponding to the first video within the second video. The method further applies area augmentation to the video signature similarity curve to determine a maximum from among a plurality of maxima corresponding to the first video file within the second video file.

US Patent:

8564721, Oct 22, 2013

Filed:

Aug 28, 2012

Appl. No.:

13/596106

Inventors:

Matthew Berry - Raleigh NC, US
Changwen Yang - Morrisville NC, US

International Classification:

H04N 9/475

US Classification:

348515, 348468, 348462, 348464, 348512, 3484254, 704252, 715203

Abstract:

The addition of temporal positions to an inverted index allows for temporal queries in addition to phrase queries. Store additional binary data for each term instance in the word-level index to prepare for searching in response to time-based queries from a user is accomplished through the use of Lucene's binary payload feature where the payload structure is defined for use in such searches. The pre-defined payload fields consist of three integers, which account for 12 extra bytes that must be stored for each term instance. A content database on the Master/Administrator server node provides the indexes for search into content in response to user events, returning results in JSON format. The search results may then be used to locate and present content segments to a user containing both requested search term results and the time location and duration within a content asset where the search term(s) is found.

US Patent:

20090094113, Apr 9, 2009

Filed:

Sep 8, 2008

Appl. No.:

12/206622

Inventors:

Matthew G. BERRY - Raleigh NC, US
Benjamin J. WEINBERGER - Durham NC, US
Schuyler E. ECKSTROM - Beaufort SC, US
Albert L. SEGARS - Beaufort SC, US

Assignee:

DIGITALSMITHS CORPORATION - Morrisville NC

International Classification:

G06Q 30/00

US Classification:

705 14

Abstract:

A system for using metadata from a video signal to associate advertisements therewith, comprising (i) a segmentation system to divide the video signal into video clips, (ii) a digitizing system for digitizing the video clips, (iii) a feature extraction system for extracting audio and video features from each video clip, associating each audio feature with respective video clips, associating each video feature with respective video clips, and saving the audio and video features into an associated metadata file, (iv) a web interface to the feature extraction system for receiving the video clips, and (v) a database, wherein video signals and associated metadata files are stored and indexed, wherein the associated metadata file is provided when a video player requests the corresponding video signal, enabling selection of a relevant advertisement for presentment in conjunction with respective video clips based on the associated audio and video features of the respective video clip.

US Patent:

20090141940, Jun 4, 2009

Filed:

Dec 3, 2008

Appl. No.:

12/327589

Inventors:

Liang ZHAO - Morrisville NC, US
Matthew G. BERRY - Raleigh NC, US

Assignee:

DIGITALSMITHS CORPORATION - Morrisville NC

International Classification:

G06K 9/00

US Classification:

382103

Abstract:

The present disclosure relates to systems and methods for modeling, recognizing, and tracking object images in video files. In one embodiment, a video file, which includes a plurality of frames, is received. An image of an object is extracted from a particular frame in the video file, and a subsequent image is also extracted from a subsequent frame. A similarity value is then calculated between the extracted images from the particular frame and subsequent frame. If the calculated similarity value exceeds a predetermined similarity threshold, the extracted object images are assigned to an object group. The object group is used to generate an object model associated with images in the group, wherein the model is comprised of image features extracted from optimal object images in the object group. Optimal images from the group are also used for comparison to other object models for purposes of identifying images.

US Patent:

20090235150, Sep 17, 2009

Filed:

Mar 17, 2009

Appl. No.:

12/405298

Inventors:

Matthew G. Berry - Raleigh NC, US

Assignee:

DIGITALSMITHS CORPORATION - Morrisville NC

International Classification:

G06F 17/00
G06F 17/30

US Classification:

715205, 707 3, 707E17014

Abstract:

The present disclosure relates to systems and methods for dynamically creating hyperlinks associated with relevant multimedia content in a computer network. A hyperlink generation module receives an electronic text file from a server. The module searches the text file to identify keywords present in the file. Once the keywords have been identified, a database is queried to identify multimedia content that is related to the keywords. Generally, multimedia content is associated with metadata to enable efficient searching of the multimedia content. Typically, the multimedia content is contextually relevant to both the identified keywords and text file. One or more hyperlinks corresponding to the keywords are then generated and inserted into the text file. The hyperlinks provide pointers to the identified multimedia content. After insertion into the text file, the hyperlinks may be clicked by a user or viewer of the file to retrieve and display the identified multimedia content.