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

A video classification method and device

A video classification and video technology, applied in the field of data processing, can solve the problems of low efficiency of classification methods and achieve the effect of improving accuracy

Active Publication Date: 2020-11-24
BEIJING WEIBOYI TECH CO LTD
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In view of this, the main purpose of the present invention is to solve the problem of low efficiency of existing classification methods

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
  • A video classification method and device
  • A video classification method and device
  • A video classification method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0023] like figure 1 As shown, the present invention provides a video classification method, comprising:

[0024] Step 101, obtain the multimodal feature vector corresponding to the video to be classified.

[0025] In this embodiment, the process of obtaining the multimodal feature vector through step 101 includes: obtaining the image features of the video to be classified; obtaining the text features of the video to be classified; fusing the image features and the text features to obtain the video corresponding to the video to be classified The multimodal feature vector of .

[0026] Among them, the way to obtain the image features of the video to be classified can be either feature extraction, or a combination of fine-tuning FineTune and feature extraction, or a classification model such as I3D / P3D / TSN network, which will not be repeated here. A repeat.

[0027] The way to obtain the text features of the video to be classified can be a modeling method such as Bag Of Words...

Embodiment 2

[0041] like figure 2 As shown, the embodiment of the present invention provides a video classification device, including:

[0042] A vector acquisition unit 201, configured to acquire a multimodal feature vector corresponding to the video to be classified;

[0043] The classification unit 202 is connected with the vector acquisition unit and the pre-trained multi-level multi-label classification model, and is used to input the multi-modal feature vector into the pre-trained multi-level multi-label classification model to obtain the hierarchical classification of the video to be classified;

[0044] Pre-trained multi-level multi-label classification model 203, including:

[0045] A global probability acquisition module that obtains the global classification probability by combining a densely connected convolutional neural network with a multi-layer perceptron;

[0046] A local probability acquisition module that obtains local classification probabilities of each layer throug...

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 discloses a video classification method and device, and relates to the field of data processing. The invention is invented to solve the problem of low efficiency of the classification method provided by the prior art. The technical solution provided by the embodiment of the present invention includes: obtaining the multimodal feature vector corresponding to the video to be classified; inputting the multimodal feature vector into a pre-trained multi-level multi-label classification model to obtain the hierarchical classification of the video to be classified The multi-level multi-label classification model includes the process of obtaining the global classification probability through the combination of the densely connected convolutional neural network and the multi-layer perceptron, the process of obtaining the local classification probability of each layer through feature feedback, and through One or more processes in the process of obtaining hierarchical classification probabilities by stacking convolutional neural networks. This scheme can be applied in the field of short video classification.

Description

technical field [0001] The invention relates to the field of data processing, in particular to a video classification method and device. Background technique [0002] In recent years, the field of short videos has developed rapidly, and various short video platforms generate a large amount of short video data every day, but to effectively use these data, it is necessary to classify and label short videos. In order to improve the timeliness and accumulation problems existing in traditional manual labeling methods, automatic classification has become a key technology that organizations in various content fields are very concerned about. [0003] In the prior art, a local method is generally used to transform the hierarchical multi-label classification problem into a prediction problem for multiple single labels according to the hierarchical structure of the category labels. By building a model for each category of the predicted output space, and then combining each model to O...

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): G06F16/75G06N3/04
CPCG06F16/75G06N3/045
Inventor 耿焕邓积杰林星白兴安徐扬
Owner BEIJING WEIBOYI TECH CO LTD