Live commenting generation method and device, electronic equipment and computer storage medium

A barrage and generator technology, applied in the computer field, can solve problems such as single style and low interest, and achieve the effect of improving interest and enriching the display effect

Active Publication Date: 2020-12-01
BEIJING CENTURY TAL EDUCATION TECH CO LTD
View PDF7 Cites 2 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 bullet chatting generation method, device, electronic equipment and computer storage

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
  • Live commenting generation method and device, electronic equipment and computer storage medium
  • Live commenting generation method and device, electronic equipment and computer storage medium
  • Live commenting generation method and device, electronic equipment and computer storage medium

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0025] Example 1,

[0026] Embodiment 1 of this application provides a barrage generation method, such as Figure 1 Shown, Figure 1 A flowchart of a barrage generation method provided by this embodiment of the application includes the following steps:

[0027] S101, acquiring a target image and text data representing emotional information.

[0028] The target image in this application can be an image containing any picture in different scenes, for example, it can be an animal photo or a test paper, etc. The embodiment of this application does not limit the specific form of the target image. Text data representing emotional information can be collected from social media software platforms using web crawlers. Social media software platforms can be comment websites, movie reviews, video barrage, live broadcast platforms, etc., and any appropriate text data with emotional information. The emotional type of the emotional information can be any one of positive, neutral and negative. Amon...

Example Embodiment

[0042] Example 2,

[0043] Based on the scheme in the first embodiment, before the emotional barrage is generated by the barrage generator in the generative confrontation neural network, the embodiment of the present application also trains the generative confrontation neural network to obtain the trained generative confrontation neural network, such as Figure 2 Shown, Figure 2 A flow chart of a training method of a barrage generator provided by this embodiment of the application is provided. The barrage generator is trained and generated by the following ways:

[0044] S201, acquiring image sample features and text sample label features.

[0045] Among them, image sample features are used to indicate the features of image samples used for training neural network models, such as generating antagonistic neural networks, while text sample label features are used to indicate the features of text samples used as labels for training neural network models, such as generating antagonisti...

Example Embodiment

[0076] Example 3,

[0077] Based on the scheme in Embodiment 2, optionally, in an embodiment of the present application, step 203a1 may include steps 203a11-203a13, as follows:

[0078] Step 203a11: Input multi-modal features to generate a barrage generator in the confrontation neural network, and generate at least one predicted word.

[0079] In this embodiment of the application, the barrage generator in the anti-neural network generated by multi-modal feature input generates at least one predicted word in turn, in which the predicted word generated later depends on the predicted word generated earlier. In a sentence, the previous predicted word has a great influence on the part-of-speech prediction of the current word. For example, when the sentence "I eat apples" is generated, when predicting apples, it is obvious that the previous eating is a verb. An emotional barrage includes at least one predicted word and one word representing emotional information.

[0080] Step 203a12: ...

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

The embodiment of the invention provides a live commenting generation method and device, electronic equipment and a computer storage medium. The live commenting generation method comprises the steps of acquiring a target image and text data representing emotion information; performing feature extraction on the target image to obtain image features, and performing feature extraction on the text data to obtain text features; and inputting the image features and the text features into a trained live commenting generator to generate an emotional live commenting corresponding to the target image. According to the invention, feature extraction is carried out on text data representing emotion information, the extracted text features include rich emotions, and then the image features and the textfeatures of the target image are input into the live commenting generator, so that the emotional live commenting generated by the live commenting generator is matched with the target image, the display effect of the emotional live commenting in the target image is enriched, and the interestingness of the emotional live commenting is improved.

Description

Technical field [0001] The embodiment of the application relates to the technical field of computers, in particular to a barrage generation method, device, electronic equipment and computer storage medium. technical background [0002] With the continuous development of artificial intelligence, adding text description to images has been applied to many technical fields, such as movie barrage, computer visual question and answer, live video, online lectures, etc. Adding text description to images or videos can bring fun and new ideas to users, and enhance the interaction between users and videos or images. [0003] However, when the barrage generator is used to generate the text description content consistent with the image according to the text information, in the prior art, only the semantic understanding part of the text information surface is considered, so that the generated text description content has a single style and low interest. Inventive content [0004] In view of t...

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
IPC IPC(8): G06K9/46H04N21/2187H04N21/435H04N21/4788H04N21/81
CPCH04N21/81H04N21/4788H04N21/435H04N21/2187G06V10/40
Inventor 葛旭日冀志龙刘霄
Owner BEIJING CENTURY TAL EDUCATION TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap