Unlock instant, AI-driven research and patent intelligence for your innovation.

Video recommendation method, device and storage medium

A recommendation method, film and television technology, applied in the electronic field, to achieve the effect of improving accuracy and reducing the difficulty of implementation

Active Publication Date: 2022-04-19
优地网络有限公司
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In view of this, the embodiment of the present application provides a video recommendation method, device and storage medium to solve the problem that existing recommendation algorithms need to rely on user behavior log information to perform video recommendation

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Video recommendation method, device and storage medium
  • Video recommendation method, device and storage medium
  • Video recommendation method, device and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0056] see figure 1 , an embodiment of the video recommendation method in the embodiment of the present application includes:

[0057] 101. Obtain the target video image;

[0058] A target video image is acquired, and the target video image is a reference image for recommending video works to the user.

[0059] Wherein, the target film and television image corresponds to a film and television work that the user is interested in. Exemplarily, in practical applications, the target video image may be an image corresponding to a video content clicked or browsed by the user, such as a movie poster.

[0060] Exemplarily, in practical applications, when a user clicks an image link corresponding to a certain film and television work while browsing a film and television page, the film and television image corresponding to the film and television work can be used as the target film and television image in the embodiment of the present application.

[0061] 102. Extracting target imag...

Embodiment 2

[0080] In order to have a clearer understanding of the content of the video recommendation method in this application, this application provides a more detailed application example for illustration, please refer to figure 2 , another embodiment of the video recommendation method in the embodiment of the present application includes:

[0081] 201. Set up a film and television image database;

[0082] Exemplarily, it is possible to extract a poster recommendation set corresponding to a film and television work from the movie resource library on the Internet (wherein, a poster or a film and television image corresponds to a film and television work), and use decentralization and nearest neighbor processing on the poster picture For the picture, standardize the corresponding picture format, and use the set of video and video images after the standardized format as the video and video image database in the embodiment of this application.

[0083] Among them, decentralization proc...

Embodiment 3

[0109] see image 3 , the embodiment of the present application also provides a video and video recommendation device implementing the above video and video recommendation method, including:

[0110] The acquisition unit 301 is configured to acquire a target video image, and the target video image is a reference image for recommending video works to users;

[0111] The extraction unit 302 is used to extract the target image features of the target video image through an image feature extraction model, the image feature extraction model being a residual network model trained in a video resource picture library;

[0112] The first calculation unit 303 is used to calculate the cosine similarity between the image features corresponding to the N candidate video images in the video image database and the target image features respectively, and obtain N cosine values, and the N is an integer greater than one;

[0113] The second calculation unit 304 is configured to calculate the sty...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

This application is applicable to the field of electronic technology, and provides a video and television recommendation method, device and storage medium. The method includes: acquiring a target video and video image, which is a reference image for recommending video and television works to users; extracting the target video and video through an image feature extraction model The target image feature of the image, the image feature extraction model is a residual network model trained through the film and television resource picture library; calculate the cosine similarity between the image features corresponding to the N candidate film and television images in the film and television image database and the target image features respectively, Obtain N cosine values, and the N is an integer greater than one; respectively calculate the style similarities between the N candidate video and video images and the target video and video images according to the N cosine values, and obtain N style similarity values; according to the N cosine values The style similarity value recommends to the user a film and television work corresponding to the candidate film and television image.

Description

technical field [0001] The present application belongs to the field of electronic technology, and in particular relates to a video recommendation method, device and storage medium. Background technique [0002] At present, video platforms basically have an intelligent video recommendation system. After users view the introduction of a certain video or watch a certain video through the video platform, the intelligent video recommendation system of the platform will automatically recommend and present the same video as the user viewed or watched. Related videos can help users find related videos without manually searching, which not only facilitates users, but also increases the user's stay time on the video website. [0003] The recommendation algorithms in the prior art all use various algorithms to analyze user behavior log information, or combine relevant movie data for correlation analysis, and the data analysis stage requires a lot of time and experience accumulation. A...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06F16/783G06F16/9535
CPCG06F16/783G06F16/9535
Inventor 谢宜廷李延平
Owner 优地网络有限公司