Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A video segment query system and video segment query method

A video clip and query system technology, applied in the field of computer vision, can solve the problems of data storage and retrieval efficiency burden, and achieve the effect of realizing personalized configuration, reducing storage space and calculation amount, and comprehensive training effect

Active Publication Date: 2021-08-03
广东众聚人工智能科技有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

On the other hand, in the related methods, the original video and query sentences are needed when positioning the video moment, and the processing of the original video often requires a large amount of calculation. Retrieval efficiency comes with a burden

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 segment query system and video segment query method
  • A video segment query system and video segment query method
  • A video segment query system and video segment query method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0068] Based on the above situation, this embodiment proposes a video segment query system. The system proposes a video event query framework (vidEo momentT loCation method via Hashing, ETCH) based on hashing's natural semantic video event query technology. The system includes:

[0069] A preprocessing module, configured to obtain a query video, wherein the query video includes T video segments; N video frames are intercepted from each video segment to obtain T×N video frames of the query video, wherein N , T is an integer greater than 1;

[0070] A video hashing network, connected to the preprocessing module, the video hashing network is used to extract video features of the query video based on the T×N video frames, and map the video features to a public space , generating a second feature vector group of the query video and a hash code of the query video;

[0071] A statement hashing network, configured to obtain a query statement, extract semantic features of the query ...

Embodiment 2

[0118] This embodiment provides a video segment query method. The method is based on the video segment query system described in Embodiment 1, and is used to predict the position of the video segment matching the query statement in the video. image 3 It is a flow chart of a video segment query method provided by an embodiment of the present invention. Such as image 3 As shown, the method includes steps S10-S40.

[0119] S10: Obtain a plurality of training pairs, wherein each training pair includes a training video and a training sentence, wherein the training video includes T training video segments, and T is an integer greater than 1; each of the training pairs is marked, marked The content of includes: a real number similarity score between each training video segment and the training sentence, and the start time and end time of the video segment matching the training sentence in the training video.

[0120] S20: Construct any video segment query system provided in Embo...

Embodiment 3

[0152] Figure 4 It is a schematic structural diagram of a computer device provided by an embodiment of the present invention. Such as Figure 4 As shown, the device includes a processor 410 and a memory 420 . The number of processors 410 may be one or more, Figure 4 A processor 410 is taken as an example.

[0153] As a computer-readable storage medium, the memory 420 can be used to store software programs, computer-executable programs and modules, such as program instructions / modules of the video segment query method in the embodiment of the present invention. The processor 410 implements any video segment query system and video segment query method provided in the above embodiments by running software programs, instructions and modules stored in the memory 420 .

[0154] The memory 420 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application program required by at least one funct...

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 segment query system and a video segment query method. The system includes: a preprocessing module for obtaining the query video; a video hash network for extracting the video features of the query video, and mapping to the public space, generating the second feature vector group of the query video and the hash of the query video code; the statement hash network is used to obtain the query statement, extract the semantic features of the query statement, and map it to the public space, and generate the first feature vector of the query statement and the hash code of the query statement; the similarity calculation module is used to calculate each The real number similarity score and the hash similarity score between a video segment and the query statement; the position prediction network is used to predict the start and end moments of the video segment matching the query statement in the query video based on the similarity score . The present invention can reduce storage space and calculation amount, and has considerable precision at the same time.

Description

technical field [0001] The embodiments of the present invention relate to the technical field of computer vision, and in particular, to a video segment query system and a video segment query method. Background technique [0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art. [0003] Natural semantic video event query technology refers to given a video and a query statement, find the start timestamp and end timestamp of the video clip that best matches the query statement. Videos can have no subtitles or sound, and queries can be in unstructured natural language. [0004] As part of the field of video understanding, video event query has a wide range of applications in entertainment, security, and creativity. For example, a goal collection can be automatically edited from a video collection; the most interesting segment can be found from a movie being watched, and the time 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): G06F16/78G06F16/787
CPCG06F16/7867G06F16/787
Inventor 丁冬睿刘新放王潇涵纪梦娟房体品魏红雷
Owner 广东众聚人工智能科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com