Supercharge Your Innovation With Domain-Expert AI Agents!

Method of ordering method from local to global video sequence fragment

An extraction method and time sequence technology, applied in the field of extracting time series segments of wonderful behaviors, can solve problems such as lack of time sequence information

Active Publication Date: 2022-08-09
NANJING UNIV OF INFORMATION SCI & TECH
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the disadvantage of this method is that it only focuses on the local and lacks the global timing information.

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
  • Method of ordering method from local to global video sequence fragment
  • Method of ordering method from local to global video sequence fragment
  • Method of ordering method from local to global video sequence fragment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0044] This embodiment adopts the local-to-global-based video time sequence segment extraction method proposed by the present invention, and the local-to-global video time sequence segment extraction system adopted by the present invention includes a feature extraction module, an encoding module, a nominated feature prediction module, and an evaluation module. module composition.

[0045] like figure 1 As shown, the inventive method comprises the following specific steps:

[0046] 1) First use the feature extraction module to extract the visual-semantic spatio-temporal features of each time point in a given video frame. The feature extraction module mainly adopts the two-stream network widely used in the video field, that is, two branches are used to model static appearance information and dynamic motion information respectively, and the inputs of the two branches are pictures and optical flow respectively. The specific process is as follows: first, the optical flow of a giv...

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 method for extracting video time series segments from local to global. The method includes: 1) feature extraction; 2) encoding; 3) nomination feature prediction; 4) evaluation; In the process, it pays attention to both the local information of the video and the global information of the video. At the same time, it can pay more attention to the behavior itself rather than the background, so as to accurately extract the fragments of the behavior in the video.

Description

technical field [0001] The invention belongs to the field of computer science artificial intelligence, and in particular relates to a method for extracting video time series segments from local to global, which is used for extracting wonderful behavior time series segments from short videos. Background technique [0002] Time-series segment nomination generation is an important part of video analysis. Its purpose is to precisely locate the time segments where actions occur in unedited videos. The quality of the generated time-series segments has a significant impact on the next step for nomination-based action detection. Great influence. How to generate clips with precise temporal boundaries is a key issue for temporal clip nomination generation. [0003] The time series segment nomination generation methods are roughly divided into the following three types: the first is a sliding window-based method. Specifically, a large number of candidate boxes of different scales are ...

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): H04N21/44H04N21/845G06N3/04
CPCH04N21/44008H04N21/8456G06N3/044G06N3/045
Inventor 朱轶昇刘光灿
Owner NANJING UNIV OF INFORMATION SCI & TECH
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com