CN114064975A - Video label generation method - Google Patents
Video label generation method Download PDFInfo
- Publication number
- CN114064975A CN114064975A CN202111419716.1A CN202111419716A CN114064975A CN 114064975 A CN114064975 A CN 114064975A CN 202111419716 A CN202111419716 A CN 202111419716A CN 114064975 A CN114064975 A CN 114064975A
- Authority
- CN
- China
- Prior art keywords
- video
- similar
- videos
- similarity
- label
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 14
- 238000001914 filtration Methods 0.000 claims abstract description 6
- 238000010606 normalization Methods 0.000 claims description 4
- 239000003016 pheromone Substances 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/7867—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, title and artist information, manually generated time, location and usage information, user ratings
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9536—Search customisation based on social or collaborative filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Library & Information Science (AREA)
- Multimedia (AREA)
- Life Sciences & Earth Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a video label generation method, which relates to the technical field of video labels, and is characterized in that based on user behavior data, a similar video set of a video and the similarity of each similar video in the video and the similar video set are obtained through a collaborative filtering algorithm or a swing algorithm, then the similarity is normalized to obtain a second similarity of the similar video, information elements of the similar video are used as initial labels of the video, the weight of each initial label is obtained through the second similarity, the initial label with the weight larger than a preset value is used as a label of the video, and the problems that the existing manual video label editing is low in accuracy and consumes a large amount of manpower and time are solved.
Description
Technical Field
The invention relates to the technical field of video tags, in particular to a video tag generation method.
Background
With the rapid development of the film and television industry, various videos are provided on the internet for people to select to watch, people can select favorite contents to watch according to the video tags, or a recommendation system recommends videos of interest for users according to the video tags. However, the tags of these videos are mostly edited manually, and since people have different understandings of objects, the tags to be played for the same video are different, and manual editing of the tags takes a lot of labor and time.
Disclosure of Invention
The technical problems solved by the invention are as follows: the video tag generation method is provided, and the problems that the existing manual video tag editing accuracy is low, and a large amount of manpower and time are consumed are solved.
The invention adopts the technical scheme for solving the technical problems that: the video label generation method comprises the following steps:
s01, acquiring a similar video set of videos and the similarity between the videos and each similar video in the similar video set;
s02, carrying out normalization processing on the similarity to obtain a second similarity of each similar video;
s03, taking information elements of similar videos as initial labels of the videos, and obtaining an initial label set of the videos;
s04, calculating the weight of each initial label by using the second similarity;
and S05, taking the initial label with the weight larger than the preset value in the initial label set as the label of the video.
Further, in step S01, the similar video set and the similarity between the video and each similar video in the similar video set are obtained through a collaborative filtering algorithm.
Further, in step S01, a swing algorithm is used to obtain the similar video set and the similarity between the video and each similar video in the similar video set.
Further, in step S03, the information elements of the similar videos include actors, director, genre and keywords.
Further, in step S04, the weight of the initial label is equal to the sum of the second similarities of all similar videos including the initial label.
The invention has the beneficial effects that: according to the video tag generation method, based on user behavior data, a similar video set of a video and the similarity between the video and each similar video in the similar video set are obtained through a collaborative filtering algorithm or a swing algorithm, then normalization processing is carried out on the similarity to obtain a second similarity of the similar video, information elements of the similar video are used as initial tags of the video, the weight of each initial tag is obtained through the second similarity, the initial tags with the weights larger than a preset value are used as tags of the video, and the problems that the accuracy of existing manual video tag editing is low, and a large amount of manpower and time are consumed are solved.
Drawings
Fig. 1 is a flow chart of a video tag generation method according to the present invention.
Detailed Description
The video tag generation method of the invention, as shown in figure 1, comprises the following steps:
s01, acquiring a similar video set of videos and the similarity between the videos and each similar video in the similar video set;
specifically, a similar video set of videos and the similarity between the video and each similar video in the similar video set are obtained through a collaborative filtering algorithm or a swing algorithm. Such as: when a certain video V is known, similar videos and similarity with the video V are found in a video library through a collaborative filtering algorithm or a swing algorithm, the obtained similar videos are arranged according to the sequence of similarity from large to small, and the similar videos with the top rank of 20 are selected as elements in a similar video set, namely the similar videos V1, V2 and V3 … V20 are obtained.
S02, carrying out normalization processing on the similarity to obtain a second similarity of each similar video;
specifically, the highest similarity among the 20 similar videos is used as a denominator, and the similarity of the similar videos is used as a numerator, so as to obtain a second similarity of each similar video in the similar video set, where the second similarity is not greater than 1.
S03, taking information elements of similar videos as initial labels of the videos, and obtaining an initial label set of the videos;
specifically, the information elements of the similar videos include actors, directors, genre and keywords.
S04, calculating the weight of each initial label by using the second similarity;
specifically, the weight of the initial tag is equal to the sum of the second similarities of all similar videos including the initial tag, for example: the similarity of the movie V to similar movies V1, V2, and V3 is 0.9, 0.8, and 0.7, respectively; v1 for the pheromones a and b, V2 for the pheromones b and c, V3 for the pheromones a and d, and V4 through V20 for none of the pheromones a, b, c and d; then, the information elements a, b, c and d are all initial labels of the video, and the weight of the initial label a is 0.9+ 0.7-1.6; the weight of the initial label b is 0.9+0.8 ═ 1.7; the weight of the initial label c is 0.8 and the weight of the initial label d is 0.7.
And S05, taking the initial label with the weight larger than the preset value in the initial label set as the label of the video.
Specifically, if the preset value is 1.5, the initial labels a and b are satisfied, and the labels of the video are a and b.
In addition, in step S01 of the present invention, the number of the similar videos selected in the similar video set is set according to the actual situation.
Claims (5)
1. The video label generation method is characterized by comprising the following steps:
s01, acquiring a similar video set of videos and the similarity between the videos and each similar video in the similar video set;
s02, carrying out normalization processing on the similarity to obtain a second similarity of each similar video;
s03, taking information elements of similar videos as initial labels of the videos, and obtaining an initial label set of the videos;
s04, calculating the weight of each initial label by using the second similarity;
and S05, taking the initial label with the weight larger than the preset value in the initial label set as the label of the video.
2. The method for generating labels of videos as claimed in claim 1, wherein in step S01, the similarity between the similar video set and each similar video in the video and similar video set is obtained through a collaborative filtering algorithm.
3. The method for generating labels of videos as claimed in claim 1, wherein in step S01, the similarity between the set of similar videos and each of the videos in the set of similar videos is obtained by swing algorithm.
4. The method for generating labels for video according to any of claims 1-3, wherein in step S03, the information elements of the similar video include actors, director, genre and keywords.
5. The method for generating labels of videos as claimed in any one of claims 1-3, wherein in step S04, the weight of the initial label is equal to the sum of the second similarities of all similar videos containing the initial label.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111419716.1A CN114064975A (en) | 2021-11-26 | 2021-11-26 | Video label generation method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111419716.1A CN114064975A (en) | 2021-11-26 | 2021-11-26 | Video label generation method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114064975A true CN114064975A (en) | 2022-02-18 |
Family
ID=80276508
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111419716.1A Pending CN114064975A (en) | 2021-11-26 | 2021-11-26 | Video label generation method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114064975A (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103049479A (en) * | 2012-11-26 | 2013-04-17 | 北京奇虎科技有限公司 | Method and system for generating online video label |
CN105404698A (en) * | 2015-12-31 | 2016-03-16 | 海信集团有限公司 | Education video recommendation method and device |
CN106407484A (en) * | 2016-12-09 | 2017-02-15 | 上海交通大学 | Video tag extraction method based on semantic association of barrages |
CN106446135A (en) * | 2016-09-19 | 2017-02-22 | 北京搜狐新动力信息技术有限公司 | Method and device for generating multi-media data label |
CN110059222A (en) * | 2019-04-24 | 2019-07-26 | 中山大学 | A kind of video tab adding method based on collaborative filtering |
-
2021
- 2021-11-26 CN CN202111419716.1A patent/CN114064975A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103049479A (en) * | 2012-11-26 | 2013-04-17 | 北京奇虎科技有限公司 | Method and system for generating online video label |
CN105404698A (en) * | 2015-12-31 | 2016-03-16 | 海信集团有限公司 | Education video recommendation method and device |
CN106446135A (en) * | 2016-09-19 | 2017-02-22 | 北京搜狐新动力信息技术有限公司 | Method and device for generating multi-media data label |
CN106407484A (en) * | 2016-12-09 | 2017-02-15 | 上海交通大学 | Video tag extraction method based on semantic association of barrages |
CN110059222A (en) * | 2019-04-24 | 2019-07-26 | 中山大学 | A kind of video tab adding method based on collaborative filtering |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111683209B (en) | Mixed-cut video generation method and device, electronic equipment and computer-readable storage medium | |
JP3568117B2 (en) | Method and system for video image segmentation, classification, and summarization | |
Cong et al. | Towards scalable summarization of consumer videos via sparse dictionary selection | |
CN101281540B (en) | Apparatus, method and computer program for processing information | |
CN108776676B (en) | Information recommendation method and device, computer readable medium and electronic device | |
US20140324845A1 (en) | Universal Lookup of Video-Related Data | |
CN110574387A (en) | Recommending live streaming content using machine learning | |
Bost et al. | Remembering winter was coming: Character-oriented video summaries of TV series | |
US20170220867A1 (en) | Identifying presentation styles of educational videos | |
CN109255037B (en) | Method and apparatus for outputting information | |
CN110287375B (en) | Method and device for determining video tag and server | |
EP1917810A1 (en) | Method and apparatus for encoding/playing multimedia contents | |
Basavarajaiah et al. | GVSUM: generic video summarization using deep visual features | |
US20130325865A1 (en) | Method and Server for Media Classification | |
CN102236714A (en) | Extensible markup language (XML)-based interactive application multimedia information retrieval method | |
CN113537215A (en) | Method and device for labeling video label | |
CN113934888A (en) | Video tag processing method and device | |
US7609761B2 (en) | Apparatus and method for abstracting motion picture shape descriptor including statistical characteristic of still picture shape descriptor, and video indexing system and method using the same | |
US8693850B2 (en) | System and method for video summarization and navigation based on statistical models | |
CN114064975A (en) | Video label generation method | |
CN105260368B (en) | A kind of editor of video data, the method for pushing of business object, device and system | |
Ritzer | Media and Genre: Dialogues in Aesthetics and Cultural Analysis | |
Klippgen et al. | The Use of Metadata for the Rendering of Personalized Video Delivery. | |
Bailer et al. | A video browsing tool for content management in postproduction | |
CN114492313B (en) | Encoder training method, resource recommendation method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |