Portrait audio video synchronous calibration device and method
A technology of voice and video, calibration device, applied in image communication, instrument, character and pattern recognition, etc., can solve problems such as inability to recognize, out of synchronization of voice information and video information, inability to judge motion without sound, etc., to improve computing power. performance, the effect of reducing the amount of information storage
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Example Embodiment
[0065] Embodiment 1: Detection process of synchronized voice and video
[0066] Step S1, read the audio and video header file information, obtain the total length of time 72 of the audio and video, the unit is seconds, and a certain moment of the audio and video is t, 1≤t≤72;
[0067] Step S2, set the dynamic lips array P[k], 1≤k≤72, set the initial value of all elements in the array P to 0, set the vocal array S[f], 1≤f≤72, set the array S All elements in the initial value are set to 0;
[0068] Step S3, sequentially extracting the picture frames of the video file at time t, image 3 is the binary image of the picture frame extracted at the 32nd second of the video file, Image 6 It is the binary image of the picture frame extracted in the 31st second of the video file, and the face recognition technology is used to identify the i face area M in the picture frame at a certain moment. t,i , 1≤i≤I, I=1, Figure 4 From image 3 A face region M extracted from 32,1 , Figur...
Example Embodiment
[0077] Example 2: Detection and Calibration Process for Asynchronous Voice and Video
[0078] Step S1, read the audio and video header file information, obtain the total length of the audio and video time 58, the unit is seconds, a certain moment of the audio and video is t, 1≤t≤58;
[0079] Step S2, set the dynamic lips array P[k], 1≤k≤58, set the initial value of all elements in the array P to 0, set the vocal array S[f], 1≤f≤58, set the array S All elements in the initial value are set to 0;
[0080] Step S3, sequentially extracting the picture frames of the video file at time t, Figure 11 is the binary image of the picture frame extracted from the 19S video file, Figure 14 It is the binary image of the picture frame extracted from the 18th second of the video file, and the face recognition technology is used to identify the i face area M in the picture frame at a certain moment. t,i , 1≤i≤I, I=3, Figure 12 From Figure 11 The three face regions M extracted from 19...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap