Browsing and retrieval of full broadcast-quality video

• US 7,877,774 B1
• Filed: 04/19/2000
• Issued: 01/25/2011
• Est. Priority Date: 04/19/1999
• Status: Expired due to Fees

• Current Assignee: AT&T Corporation (AT&T, Inc.)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: Gibbon, David Crawford, Shahraray, Behzad, Civanlar, Mehmet Reha, Pieraccini, Roberto, Basso, Andrea, Huang, Qian, Levin, Esther
• Primary Examiner(s): Shang; Annan Q
• Application Number:
  US09/552,312
• Time in Patent Office: 3,933 Days
• Field of Search: 725/92, 725/115, 725/139, 725/144, 725 37- 61, 386/69, 386/46, 386/56, 386/92, 386/95, 386/1, 386/4, 360/13, 360/32
• US Class Current: 725/44
• CPC Class Codes: G06F 16/41 Indexing; Data structures therefor; Storage structures
• CPC Class Codes: H04N 21/47202 for requesting content on demand, e.g. video on demand
• CPC Class Codes: H04N 21/4828 for searching program descriptors retrieval of video data G06F16/739
• CPC Class Codes: H04N 21/8456 by decomposing the content in the time domain, e.g. in time segments

Abstract

A method includes steps of indexing a media collection, searching an indexed library and browsing a set of candidate program segments. The step of indexing a media collection creates the indexed library based on a content of the media collection. The step of searching the indexed library identifies the set of candidate program segments based on a search criteria. The step of browsing the set of candidate program segments selects a segment for viewing.

Claims

1. A method comprising steps of:

classifying a media collection comprising program content and television commercials as classified program content versus classified television commercials, wherein the television commercials are presented chronologically separate from the program content and in a same display screen as the program content;

identifying program segments within the classified program content based on synchronizing recognized speech in each identified segment with captioning to extract stories and to yield synchronized speech, wherein synchronizing recognized speech comprises;

recognizing speech in the media collection to yield recognized speech,

generating word timestamps based on the recognized speech, and

aligning the captioning with the recognized speech based on the word timestamps;

indexing the media collection to create an indexed library based on the program segments to yield an indexed media collection;

receiving at a server a search query to the indexed media collection from a user;

searching the indexed library to identify a set of candidate program segments based on the search query; and

presenting at a client device of the user the set of candidate program segments for the user to browse and select.

2. The method of claim 1, wherein:

the step of indexing further includes a step forming a browseable image for each segment of the set of candidate program segments, each browseable image including keywords identified in the searchable text data for display in the browseable image; and

the step of presenting includes selecting a display segment from the set of candidate program segments and displaying the associated browseable image with associated keywords.

3. The method of claim 2, wherein:

each browseable image further includes key images identified in the indexed library for display in the browseable image; and

the step of displaying the associated browseable image further comprises displaying associated key images.

4. The method of claim 2, wherein:

the searchable text data associated with the selected display segment includes a first word having low information content and a second word having high information content; and

the step of forming a browseable image includes selecting the second word as a keyword and rejecting the first word as a keyword.

5. The method of claim 1, wherein:

the step of indexing further includes a step forming a browseable image for each segment of the set of candidate program segments, each browseable image including key images identified in the indexed library for display in the browseable image; and

the step of browsing includes selecting a display segment from the set of candidate program segments and displaying the associated browseable image.

6. The method of claim 5, wherein:

the media associated with the selected display segment includes an image of an anchor-person having low information content and a field shot image of an event having high information content; and

the step of forming a browseable image includes selecting the field shot image as a key image and rejecting the image of the anchor-person as a key image.

7. A system for video indexing and delivery, the system comprising:

a first module configured to classify video into program segments versus television commercial segments, wherein the television commercials are presented chronologically separate from the program content and in a same display screen as the program content;

a second module configured to identify speaker segments within the program segments based on speaker voice characteristics to yield identified speaker segments;

a third module configured to extract stories from the identified speaker segments based on synchronizing speech with closed captioning of the identified speaker segments, the synchronizing based on;

recognizing speech in the video,

generating word timestamps based on the recognized speech, and

aligning the captioning with the recognized speech based on the word timestamps;

a fourth module configured to receive a natural language query;

a fifth module configured to select key frames from identified speech segments in response to the query; and

a sixth module configured to present the key forms to a user device for browsing by the user.