Video auditing method based on multiple levels and multiple models, medium and computer equipment

A multi-model, multi-level technology, applied in the computer field, can solve problems such as increased load on online machines, increased engineering difficulty, and increased online pressure

Active Publication Date: 2020-07-07
BIGO TECH PTE LTD
View PDF7 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, using too many models will increase engineering difficulty and online pressure,...

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 auditing method based on multiple levels and multiple models, medium and computer equipment
  • Video auditing method based on multiple levels and multiple models, medium and computer equipment
  • Video auditing method based on multiple levels and multiple models, medium and computer equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0048]Embodiments of the present invention are described in detail below, examples of which are shown in the drawings, wherein the same or similar reference numerals designate the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the figures are exemplary only for explaining the present invention and should not be construed as limiting the present invention.

[0049] Such as figure 1 Shown, in one embodiment, a kind of video review method based on multi-level multi-model, comprises:

[0050] S11. Input the video picture obtained after processing the video stream into the first-level review model to obtain a first violation score.

[0051] In the embodiment of the present invention, the video picture obtained after processing the video stream is input to the first-level review model, and before the first violation score is obtained, the video stream needs to be segmented according to a preset s...

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 invention provides a video auditing method and device based on multiple levels and multiple models, a computer readable storage medium and computer equipment, and the method comprises the steps: inputting a video picture obtained after a video stream is processed into a first-level auditing model, and obtaining a first violation score; inputting the video picture of which the first violation score is greater than a first preset threshold into a second-level auditing model to obtain a second violation score; processing the second violation score according to a preset machine learning algorithm to obtain a final violation score of the video stream, and outputting an unqualified auditing result of the video stream when judging that the final violation score is higher than a second presetthreshold. According to the invention, a multi-level and multi-model structure and an xgboost fitting score mode are adopted, so the redundant overhead is reduced while multiple models are used, and the online video auditing efficiency is improved efficiently.

Description

technical field [0001] The present invention relates to the field of computers, in particular, the present invention relates to a multi-level and multi-model-based video review method, medium and computer equipment. Background technique [0002] Video review refers to screening and filtering some illegal videos. For example, some videos involving violence, terrorism and pornography are illegal videos. At present, the solution to the problem of low-quality video review is usually: process each frame of video data, input the image data of each frame and the corresponding label (violation or non-violation) of the frame to the deep neural network, and use the deep neural network to fit the violations respectively. Features of video data and normal video data. After getting the fitted model, you can input a video to the model, and the output is the violation probability of the video. [0003] However, there are at least the following defects in the existing technical solutions:...

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): H04N21/234H04N21/44H04N21/462H04N21/454G06K9/00G06K9/62
CPCH04N21/23418H04N21/44008H04N21/462H04N21/454G06V20/41G06F18/22G06F18/2148
Inventor 杨轩刘振强
Owner BIGO TECH PTE 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