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

Short video classification method, system and medium based on multimodal dictionary learning

A dictionary learning and modal classification technology, applied in character and pattern recognition, instruments, computing, etc., can solve problems such as short videos and short videos

Active Publication Date: 2020-07-07
SHANDONG UNIV
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although some previous methods have devoted to multimodal data fusion, how to model the correlation between multimodalities and fuse them effectively is still an open research problem.
In addition, short videos have their outstanding and unique features: 1) Short videos are relatively short and usually only convey one or a few advanced topics or concepts

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
  • Short video classification method, system and medium based on multimodal dictionary learning
  • Short video classification method, system and medium based on multimodal dictionary learning
  • Short video classification method, system and medium based on multimodal dictionary learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0045] It should be pointed out that the following detailed description is exemplary and intended to provide further explanation to the present application. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0046] It should be noted that the terminology used here is only for describing specific implementations, and is not intended to limit the exemplary implementations according to the present application. As used herein, unless the context clearly dictates otherwise, the singular is intended to include the plural, and it should also be understood that when the terms "comprising" and / or "comprising" are used in this specification, they mean There are features, steps, operations, means, components and / or combinations thereof.

[0047] Given a set of labeled short videos (the scale does not need to be very large) in the initial offline sta...

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 short video classification method, system and medium based on multi-modal dictionary learning, and applies it to the short video scene prediction problem, including the following steps: Offline part: constructing a tree-shaped guided multi-mode based on a small number of labeled samples Modal dictionary learning, get the initialized multi-modal dictionary and the corresponding sparse representation; based on the learned multi-modal sparse representation, train a short video scene classifier for each modality; online part: real-time for newcomers Short video samples for scene prediction or reinforcement dictionary learning model. For new short videos, if there is a scene mark, it is used to strengthen the dictionary learning model; if the scene is not marked, the classifier trained and learned offline is used to predict the scene category.

Description

technical field [0001] The present invention relates to the technical field of short video classification, in particular to a short video classification method, system and medium based on multimodal dictionary learning. Background technique [0002] We live in an era of constantly shrinking attention time. In order to satisfy our thirst for fast content, in addition to short texts such as tweets and Weibo, short videos with "shorter-better" characteristics, with the rise of sharing services And more and more popular. Typical services include foreign Vine, Snapchat, Viddy and MixBit, domestic platforms such as Douyin, Kuaishou and Huanhan, which usually limit video length to 6, 10, 15 or 16 seconds. With the characteristics of simplicity, authenticity, dissemination and low-cost value, short videos are shaking and taking over the content and social media marketing space. [0003] As the number of short videos grows at an exponential rate, the category information of short v...

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): G06K9/62
CPCG06F18/24G06F18/214
Inventor 刘萌聂礼强郑晓云罗旺王翔宋雪萌甘甜陈宝权
Owner SHANDONG UNIV