Automated Video Title and Keyframe Generation via Subtitle Analysis

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The increasing number of content providers and consumers in the Internet TV market makes manual generation of video titles and keyframes inefficient, limiting content exposure.

Innovation Solution

A method for automatically generating video titles and keyframes by analyzing subtitles, selecting a main subtitle, extracting content information from keyframes, and combining metadata to create a title, using techniques like morphemic analysis, TextRank, and kernel PCA.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual generation of video titles and keyframes is used, then quality and accuracy are improved, but productivity and cost efficiency deteriorate

Engineering Contradiction:

Improvetitle and keyframe qualityVSAvoidgeneration speed

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The system performs automatic self-generation of titles and keyframes by analyzing video subtitles and frames through computational algorithms, eliminating the need for manual human intervention while maintaining quality standards

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

Manual mechanical processes of title creation and frame selection are replaced with automated computational systems that use morphemic analysis, TextRank, and kernel PCA algorithms to generate titles and select keyframes

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If manual selection of keyframes is used, then accuracy is improved, but loss of time increases

Engineering Contradiction:

Improvekeyframe selection accuracyVSAvoidtime for title and keyframe generation

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary analysis of video subtitles and frames before final selection, using morphemic analysis to identify main subtitles and TextRank to pre-rank potential keyframes, thereby accelerating the final selection process

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

Manual time-consuming frame-by-frame review is replaced with automated computational algorithms that efficiently identify and select optimal keyframes based on analyzed video content and metadata

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If automated generation is used, then productivity is improved, but manufacturing precision deteriorates

Engineering Contradiction:

Improvegeneration speedVSAvoidtitle and keyframe quality

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The system incorporates feedback mechanisms where generated titles and keyframes are evaluated against video content characteristics, allowing iterative refinement and adjustment to maintain high quality while preserving automated generation speed

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system adjusts generation parameters dynamically based on video characteristics, using morphemic analysis depth, TextRank weighting factors, and kernel PCA parameters to optimize both speed and quality for different content types

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS10795932B2Method and apparatus for generating title and keyframe of video

Publication Date: 2020.10.06 ELECTRONICS & TELECOMM RES INST

AI summary

Disclosed is a method and apparatus for generating a title and a keyframe of a video. According to an embodiment of the present disclosure, the method includes: selecting a main subtitle by analyzing subtitles of the video; selecting the keyframe corresponding to the main subtitle; extracting content information of the keyframe by analyzing the keyframe; generating the title of the video using metadata of the video, the main subtitle, and the content information of the keyframe; and outputting the title and the keyframe of the video.