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

Fabricated component construction monitoring video positioning method based on natural language query

A construction monitoring and natural language technology, applied in video data indexing, video data retrieval, instruments, etc., can solve problems such as inability to quickly complete video segment positioning and increased calculation

Pending Publication Date: 2022-07-22
SHANDONG JIANZHU UNIV
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Deep learning is a commonly used tool. However, when using deep learning tools to locate video clips, complex feature extractors are used to extract a large number of video and text features.
As a result, the amount of calculation increases, and the positioning of video clips cannot be quickly completed.

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
  • Fabricated component construction monitoring video positioning method based on natural language query
  • Fabricated component construction monitoring video positioning method based on natural language query
  • Fabricated component construction monitoring video positioning method based on natural language query

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0044] Attached below figure 1 and attached figure 2 The present invention will be further described.

[0045] A construction monitoring video positioning method for fabricated components based on natural language query, the specific steps are:

[0046] (a) Using a video encoder to convert the construction monitoring video of fabricated components into a feature sequence , for the first video features at each time step, is the total number of divided video time steps.

[0047] (b) Mining feature sequences After the timing information of the video in the video is generated using a fully connected layer with an activation function video hash code at time steps .

[0048] (c) Use the Glove model to The query sentence of each word is used to extract the feature operation to obtain the feature sequence. , for the first feature vector of words.

[0049] (d) Considering the different importance of words in the query sentence, the present invention uses the s...

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 fabricated component construction monitoring video positioning method based on natural language query, relates to the technical field of video image processing, and aims to quickly and accurately position a construction site segment to be searched by taking a natural language statement as a query intention of a user. A video clip positioning task based on natural language query essentially belongs to the field of cross-modal retrieval, videos and texts can be mapped to the same Hamming space through Hash learning, and different modal data are mapped to a joint embedding space for interaction and analysis, so that more effective feature information is obtained. Meanwhile, a soft attention mechanism is used for weighting query information, key vocabularies in query content are emphasized, and better performance is achieved in the modal interaction stage.

Description

technical field [0001] The invention relates to the technical field of video image processing, in particular to a method for locating construction monitoring video of prefabricated components based on natural language query. Background technique [0002] The new prefabricated building is an important part of green intelligent construction and the new construction industry. In the field construction of prefabricated components, there are problems such as difficulty in precise assembly, poor quality of the connection area, and low degree of autonomy of the support system. Therefore, the construction of prefabricated buildings is realized. Green, efficient and intelligent on-site is an urgent problem to be solved. Among them, the intelligent monitoring and analysis of the prefabricated construction site is very important for the timely detection of abnormal problems in the construction project and the improvement of assembly efficiency. Usually, the construction period of pref...

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): G06F16/71G06F16/732G06F16/783
CPCG06F16/71G06F16/7343G06F16/783
Inventor 聂秀山谭智方于德湖刘新锋刘兴波
Owner SHANDONG JIANZHU UNIV