Video Summary Generation Using Audience Review Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current video summary methods fail to accommodate individual viewer preferences, as they are pre-defined and do not account for varying viewer interests, leading to summaries that may be too long or too short for specific viewing times.

Innovation Solution

A method and system that generate video summaries based on audience reviews, allowing viewers to request summaries of specific lengths by selecting frames from the video that reflect interesting points, using statistical analysis to identify and rank candidate frames based on similarity and importance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If pre-defined video summary methods are used, then the summary generation process is simple, but the summary cannot accommodate individual viewer preferences and varying viewing time requirements

Engineering Contradiction:
Improveadaptability to viewer preferencesVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system pre-generates multiple video summaries with different lengths (e.g., 1-minute, 3-minute, 5-minute versions) in advance using statistical analysis of audience reviews. When a viewer requests a summary, the system simply retrieves the pre-generated summary of the requested length, avoiding real-time generation complexity while maintaining adaptability to different viewing time preferences.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system varies the length parameter of video summaries to accommodate different viewer preferences and viewing time constraints. By generating summaries in multiple length versions and allowing viewers to select their preferred duration, the system adapts to individual needs without requiring complex real-time adjustment mechanisms.

Inventive Principle:
Principle #35Parameter changes

2Duration of action of moving object

If traditional video summary methods are used, then the processing time is short, but the summary length does not match specific viewing time requirements

Engineering Contradiction:
Improvevideo summary lengthVSAvoidtime loss
Core Design Contradiction:
Duration of action of moving objectVSLoss of time

Solution Approach 1:

The system pre-generates video summaries in multiple length versions (1-minute, 3-minute, 5-minute, etc.) before viewer requests. This allows the system to immediately retrieve and deliver the appropriate summary length without real-time processing delays, matching viewer time constraints while minimizing actual viewing time loss.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system dynamically provides different summary lengths based on viewer requests and time constraints. By maintaining multiple pre-generated versions with varying durations, the system adapts to different viewing scenarios (e.g., quick review vs. detailed overview) without requiring complex real-time generation, thus respecting viewer time while providing appropriate content length.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS12010405B2Generating video summary
Publication Date: 2024.06.11 INTERNATIONAL BUSINESS MACHINE CORPORATION
  • US12010405B2 patent drawing
  • US12010405B2 patent drawing
  • US12010405B2 patent drawing

AI summary

A computer-implemented method includes receiving a viewer request for playing a video summary of a video, wherein the viewer request includes a length of the video summary, generating the video summary of the viewer-requested length comprising a set of frames selected from the video based on audience reviews of the video, and playing a video stream of the video summary.