# US 2019/0095529 A1 · Method and apparatus for generating title and keyframe of video

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2019/0095529 A1
- Document ID: US20190095529A1
- Status: Published application
- Kind: A1
- Application number: 16/122,621
- Filed: 2018-09-05
- Published: 2019-03-28
- Earliest claimed: 2017-09-28
- Assignees: Electronics and Telecommunications Research Institute
- Applicants: Electronics and Telecommunications Research Institute
- Inventors: Jeong Woo SON; Sun Joong KIM; Won Joo PARK; Sang Yun LEE
- Web page: https://patentgrep.com/patent/US20190095529A1

## CPC

- G06F 16/41: Indexing; Data structures therefor; Storage structures
- G06F 16/487: using geographical or spatial information, e.g. location
- G06F 16/489: {using time information}
- G06F 16/739: {in form of a video summary, e.g. the video summary being a video sequence, a composite still image or having synthesized frames}
- G06F 40/216: using statistical methods
- G06V 20/46: {Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames}
- G06V 20/47: {Detecting features for summarising video content}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

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.

## Claims

### Claim 1

A method of generating a title and a keyframe of a video, the method comprising:
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.

### Claim 2 (depends on 1)

The method of claim 1, wherein the selecting of the main subtitle by analyzing the subtitles of the video comprises:
performing a morphemic analysis on a per-morpheme basis on the subtitles of the video;
calculating a weighting of a relation between words on the basis of a result of the morphemic analysis;
calculating a relation-based word weighting on the basis of the weighting of the relation between words;
calculating weightings for respective subtitles using the relation-based word weighting; and
selecting the subtitle with a highest weighting as the main subtitle on the basis of the weightings for respective subtitles.

### Claim 3 (depends on 2)

The method of claim 2, wherein the relation-based word weighting is calculated by applying TextRank or kernel principal component analysis (kernel PCA) technique on the basis of information on the weighting of the relation between words.

### Claim 4 (depends on 1)

The method of claim 1, wherein at the selecting of the keyframe corresponding to the main subtitle, the keyframe is selected by performing at least one of motion analyzing, person detection, resolution measurement, and background sound detection on the video of a particular segment corresponding to the main subtitle.

### Claim 5 (depends on 4)

The method of claim 4, wherein the selecting of the keyframe corresponding to the main subtitle comprises:
reinforcing a frame by adjusting contrast and brightness of the selected keyframe.

### Claim 6 (depends on 1)

The method of claim 1, wherein the content information of the keyframe comprises at least one piece of information on who, where, when, what, and how.

### Claim 7 (depends on 1)

The method of claim 1, wherein the metadata of the video comprises at least one piece of information on an actor, a place, and an era of the video.

### Claim 8 (depends on 1)

The method of claim 1, wherein at the generating of the title of the video using the metadata of the video, the main subtitle, and the content information of the keyframe, the title of the video is generated by combining the content information of the keyframe with information on the metadata of the video and by adding the main subtitle to the combined content information.
