Commercial video summaries using crowd annotation

Patent metadata

Field	Value
Publication number	US-11620334-B2
Application number	US-201916686288-A
Country	US
Kind code	B2
Filing date	Nov 18, 2019
Priority date	Nov 18, 2019
Publication date	Apr 4, 2023
Grant date	Apr 4, 2023

Abstract

A system for executing a video summary is provided. One or more video segments for a video based on one or more digital media is generated. A video summary is generated based on a user request.

First claim

What is claimed is:

1. A computer-implemented method, the method comprising:
analyzing, by one or more processors, one or more digital media data to produce an analysis, wherein the one or more digital media data includes a video and a crowd-based interaction associated with the video, wherein the crowd-based interaction includes a comment;
generating, by one or more processors, one or more video segments for the video based on the one or more digital media data; and
generating, by one or more processors, a video summary associated with the video based on a user request.

2. The computer-implemented method of claim 1 , the method further comprising:
receiving, by the one or more processors, a user request;
executing, by the one or more processors, a query on a database based on the user request;
identifying, by the one or more processors, one or more digital media data associated with the user request; and
receiving, by the one or more processors, the one or more digital media data.

3. The computer-implemented method of claim 1 , the method further comprising:
analyzing, by the one or more processors, the one or more digital media data;
identifying, by the one or more processors, data associated with the one or more digital media data that includes (i) one or more videos, (ii) one or more crowd-based comments, and (iii) one or more video direct accesses;
generating, by the one or more processors, an analysis result by analyzing the one or more digital media data using one or a combination of (i) natural language processing, (ii) word2vec, (iii) cognitive AI processing, (iv) video processing, and (v) machine vision; and
identifying, by the one or more processors, one or more user feedback associated with the one or more digital media data based on the analysis result.

4. The computer-implemented method of claim 1 , the method further comprising:
analyzing, by the one or more processors, one or more user feedback;
identifying, by the one or more processors, that the one or more user feedback reaches a threshold value of similarity to the user request; and
determining, by the one or more processors, that the one or more user feedback reaches a threshold value of similarity based on a result of a semantic analysis of (i) the user request and (ii) the one or more user feedback.

5. The computer-implemented method of claim 3 , the method further comprising:
generating, by the one or more processors, one or more video segments based on the user feedback associated with the one or more digital media data; and
generating, by the one or more processors, an aggregate of the one or more video segments, wherein the aggregate includes a video summary direct access that is associated with a specific timestamp at which a given video segment begins.

6. The computer-implemented method of claim 5 , the method further comprising:
generating, by the one or more processors, the video summary which includes (i) the aggregate of the one or more video segments, and (ii) the one or more digital media data;
generating, by the one or more processors, one or more labels for the video summary based on (i) an analysis of a crowd-based comment that points to a video segment and (ii) on one or a combination of wherein the one or more labels are generated based on one or a combination of words, hash representation, or n-gram structures; and
generating, by the one or more processors, a ranking for the one or more video segments based on a calculated score of the content of the one or more crowd-based comments, wherein the given calculated score is based on a feedforward neural network used to determine a weighting factor associated with features associated with the one or more video segments.

7. The computer-implemented method of claim 5 , the method further comprising:
playing, by the one or more processors, the video summary for one or more users;
receiving, by the one or more processors, a user activity associated with the one or more users;
analyzing, by the one or more processors, the user activity; and
updating, by the one or more processors, the one or more video segments associated with the video summary based on the view time of the one or more video segments.

8. The computer-implemented method of claim 1 , the method further comprising:
identifying, by the one or more processors, one or more crowd-based interactions associated with the one or more videos;
generating an analysis result by analyzing, by the one or more processors, the one or more crowd-based interactions that include one or a combination of (i) crowd-based interaction with the one or more videos, (ii) one or more crowd-based comments associated with one or more segments of the one or more videos, or (iii) one or more crowd-based reactions associated with the one or more videos; and
determining, by the one or more processors, the one or more video segments based on the analysis result.

9. A computer program, the computer program product comprising:
one or more computer-readable storage media and program instructions stored on the one or more computer-readable storage media, the program instructions comprising:
program instructions to analyze one or more digital media data to produce an analysis, wherein the one or more digital media data includes a video and a crowd-based interaction associated with the video, wherein the crowd-based interaction includes a comment;
program instructions to generate one or more video segments for the video based on the one or more digital media data; and
program instructions to generate a video summary associated with the video based on a user request.

10. The computer program product of claim 9 , the program instructions further comprising:
program instructions to receive a user request;
program instructions to execute a query on a database based on the user request;
program instructions to identify one or more digital media data associated with the user request; and
program instructions to receive the one or more digital media data.

11. The computer program product of claim 9 , the program instructions further comprising:
program instructions to analyze the one or more digital media data;
program instructions to identify data associated with the one or more digital media data that includes (i) one or more videos, (ii) one or more crowd-based comments, and (iii) one or more video direct accesses;
program instructions to generate an analysis result by analyzing the one or more digital media data using one or a combination of (i) natural language processing, (ii) word2vec, (iii) cognitive AI processing, (iv) video processing, and (v) machine vision; and
program instructions to identify one or more user feedback associated with the one or more digital media data based on the analysis result.

12. The computer program product of claim 9 , the program instructions further comprising:
program instructions analyze one or more user feedback;
program instructions to identify that the one or more user feedback breaches a threshold value of similarity to the user request; and
program instructions to determine that the one or more user feedback reaches a threshold value of similarity based on a result of a semantic analysis of (i) the user request and (ii) the one or more user feedback.

13. The computer program product of claim 11 , the program instructions further comprising:
program instructions to generate one or more video segments based on the user feedback associated with the one or more digital media data; and
program instructions to generate an aggregate of the one or more video segments, wherein the aggregate includes a video s

Assignees

• IBM

Inventors

• Cannon Gray
• Baughman Aaron K
• Stelmar Netto Marco Aurelio
• Figueredo De Santana Vagner

Classifications

• G06N3/09

  Supervised learning · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• H04N21/4788

  communicating with other users, e.g. chatting {(arrangements for providing for computer conferences, e.g. chat rooms, to substation in data switching networks H04L12/1813; distributed application using peer-to-peer [P2P] networks H04L67/104)} · CPC title

• H04N21/47202

  for requesting content on demand, e.g. video on demand · CPC title

• G06N3/04

  Architecture, e.g. interconnection topology · CPC title