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

Video sequence information mining system and method and application thereof

A technology of information mining and video sequences, applied in special data processing applications, digital data information retrieval, instruments, etc., can solve problems such as difficulty in extracting video sequence information, achieve richness and accuracy, improve efficiency, facilitate use and The effect of maintenance

Pending Publication Date: 2021-10-22
TIANJIN UNIV OF TECH & EDUCATION TEACHER DEV CENT OF CHINA VOCATIONAL TRAINING & GUIDANCE
View PDF4 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to provide a video sequence information mining system based on deep learning to solve the problem of difficulty in extracting video sequence information in the prior art

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 sequence information mining system and method and application thereof
  • Video sequence information mining system and method and application thereof

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0031] A video sequence information mining system based on deep learning, including a capture module, a data preprocessing module, a deep learning model training module, and an identification and processing module, in which:

[0032] The capture module uses the keyword "public figure" to capture image samples through a search engine, and uses a rectangular frame to mark the face area in the captured information, and the data preprocessing module is used to process the marked image samples Perform filtering and standardization to obtain a training set and a verification set. The training model in the deep learning model training module is trained through the training set, and then the identification model is obtained through verification of the verification set. The identification and processing module is used to extract the key points in the video to be processed. Frame and record the time point, after the key frame is standardized, it is passed to the recognition model to iden...

Embodiment 2

[0036] A video sequence information mining method based on deep learning, comprising the following steps:

[0037] Step 1. The capture module uses the keyword "public figure" to capture image samples through search engines, and then supplements them with image samples from public datasets to obtain a collection of image samples. Use a rectangular frame to identify the face area in each image sample. For labeling, use the pixel coordinates of the upper left corner and the pixel coordinates of the lower right corner to represent the rectangular frame;

[0038] Step 2, the data preprocessing module filters and standardizes the labeled image samples to obtain a training set and a verification set;

[0039] Step 3, the training set trains the deep learning model in the deep learning model training module, and then uses the verification set to verify the deep learning model, saves the deep learning model with the best effect on the verification set, and obtains the recognition model...

Embodiment 3

[0048] This embodiment illustrates the excavation system in embodiment 1 or the excavation method in embodiment 2 by way of example.

3.1

[0049] Embodiment 1 or Embodiment 2 can be applied to video material interception. For example, only a segment of a certain actor is intercepted in a movie. The start and end time points of a video clip.

[0050] Video clip start mark: the time from the last recognition time of the person exceeds the threshold b seconds; video clip end mark: the time from the next time the person is recognized exceeds the threshold b seconds. For example b=30. After obtaining video clips, secondary processing and creation can be performed, saving the workload of intercepting videos.

3.2

[0051] Embodiment 1 or embodiment 2 can be applied to short video feature extraction, such as calculating which main actors are included in this short video, obtain as figure 1 The results shown are used as features in video recommendation or search. For example, t...

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 sequence information mining system and method and application thereof. In the mining system, a capturing module captures a picture sample through a search engine by means of a public character keyword and marks a face area in captured information by means of a rectangular frame; the data preprocessing module is used for filtering and standardizing marked picture samples to obtain a training set; a training model in the deep learning model training module is trained through the training set to obtain an identification model; and the identification and processing module is used for extracting key frames in a video to be processed and recording time points, and after the key frames are standardized, transmitting the key frames to an identification model to identify the public characters, and outputting the public characters contained in the key frames at each time point. The invention provides a quick and effective tool and method for further processing and processing the video.

Description

technical field [0001] The invention relates to the technical field of video processing, in particular to a video sequence information mining system, method and application thereof. Background technique [0002] With the rapid development of technologies such as deep learning, 5G, and chips, the demand for video information analysis continues to emerge. In the field of toC, the user scale of short video and small video products is increasing, and a large amount of video data needs to be analyzed and processed. Authors of short or small videos need low-cost interception of video materials, while video product platforms need to effectively analyze video results. At present, there is no public technology or platform for customized extraction of video sequence information. Contents of the invention [0003] The purpose of the present invention is to provide a video sequence information mining system based on deep learning to solve the problem of difficulty in extracting vide...

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 Applications(China)
IPC IPC(8): G06F16/735G06K9/00G06K9/62
CPCG06F16/735G06F18/217G06F18/214
Inventor 秦迎梅门聪车艳秋韩春晓
Owner TIANJIN UNIV OF TECH & EDUCATION TEACHER DEV CENT OF CHINA VOCATIONAL TRAINING & GUIDANCE
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