Supercharge Your Innovation With Domain-Expert AI Agents!

Character expression posture lie detection method and system based on deep learning

A deep learning and detection method technology, applied in the field of deep learning, can solve problems such as time-consuming and labor costs, timeliness and accuracy need to be improved, mixed with subjective judgments, understanding deviations, etc., to reduce equipment costs and improve measurement The effect of lie efficiency

Pending Publication Date: 2021-11-19
王剑
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This method consumes a lot of time and labor costs, and the conventional way of judging the authenticity of language is mixed with many subjective factors such as subjective judgment and understanding deviation, which makes the efficiency of letter collection low, and the timeliness and accuracy need to be improved

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
  • Character expression posture lie detection method and system based on deep learning
  • Character expression posture lie detection method and system based on deep learning
  • Character expression posture lie detection method and system based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0046] see figure 1 , figure 1 A schematic diagram of the steps of a method for detecting a character's expression posture and lie based on deep learning provided by the embodiment of the present invention, which is as follows:

[0047] Step S100, extracting text from the training video;

[0048] In some embodiments, according to the path of the video file, the path of the batch processing text file is obtained, according to the path as described, all text files are traversed, the vocabulary obtained is added to target_vocabulary, and the target vocabulary is generated, and the sorted function is used to sort the The target vocabulary is sorted alphabetically;

[0049] According to the obtained target vocabulary, obtain the parameter index corresponding to each vocabulary, read each sentence, and add the index parameter to the vector vect corresponding to the sentence;

[0050] List the file name, sentence vector, and trusted label as a table to get text_data, and compress ...

Embodiment 2

[0074] Such as figure 2 As shown, the text feature analysis part of the present invention mainly extracts, features compresses, features extracts and features processes the sentences described by the person under test. The main steps are to read the preprocessed text data Text_Dataset.pkl, and use the Embedding network layer to embed the text features, and map the high-dimensional original data to the low-dimensional manifold. Preferably, the compressed data dimension is 300. Dimensional change is performed on the mapped low-dimensional manifold data, and one-dimensional and two-dimensional are swapped. Finally, three one-dimensional convolutional neural networks are used to perform feature parallel extraction on low-dimensional manifold data. Preferably, the input dimension of the first one-dimensional convolutional neural network is set to 300, the output dimension is 20, and the convolution kernel size is 3. Preferably, the input dimension of the first one-dimensional co...

Embodiment 3

[0082] see Figure 7 , Figure 7 A schematic diagram of a character expression posture lie detection system module based on deep learning provided by the embodiment of the present invention, which is as follows:

[0083] Extract text module 10, be used to extract text to training video;

[0084] The conversion module 20 is used to process the extracted text and convert it into a word vector;

[0085] Preprocessing module 30, for carrying out sound extraction to training video, and generate pre-training sound feature extraction network, carry out picture extraction to training video, frame by frame segmentation, image preprocessing after segmentation;

[0086] Feature extraction module 40, for using text feature neural network to extract word vector feature, use sound feature neural network to extract sound feature, use image feature neural network to extract image feature;

[0087] Classification module 50 is used for combining the extracted sound, text, image feature vecto...

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 provides a character expression posture lie detection method and system based on deep learning, and relates to the field of deep learning. The character expression posture lie detection method based on deep learning comprises the following steps: extracting a text from a training video; processing the extracted text, and converting the text into word vectors; performing sound extraction on the training video, generating a pre-training sound feature extraction network, performing picture extraction and frame-by-frame segmentation on the training video, and processing segmented images; extracting word vector features by using a text feature neural network, extracting sound features by using a sound feature neural network, and extracting image features by using an image feature neural network; combining the extracted feature vectors of the sound, the text, and the image with action detail features, and carrying out classifying by using a classification network. In addition, the invention also provides a character expression posture lie detection system based on deep learning.

Description

technical field [0001] The present invention relates to the field of deep learning, in particular, to a method and system for detecting lies in character expression gestures based on deep learning. Background technique [0002] Intelligent polygraph analysis system is the inevitable product of the development of electronics, computer, automation and other technologies to a certain extent. Many intelligent audio and video analysis systems have become effective tools for solving the credibility of evidence in our country. In recent years, my country has intensively issued a number of policies to support the development of computer vision technology and speech recognition technology. As my country's computer vision technology and speech recognition technology tend to mature. [0003] The traditional language authenticity judgment is mainly through manual recording of text information, and multiple rounds of manual review and verification to confirm whether what the speaker said...

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
IPC IPC(8): G06K9/00G06F40/151G06K9/34G06K9/46G06K9/62G06N3/04G10L17/18G10L17/26
CPCG06F40/151G10L17/26G10L17/18G06N3/045G06F2218/08G06F18/2411G06F18/253
Inventor 王健祝偲博皖彦淇张嘉琪李拓周为
Owner 王剑
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