<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>The reliance on shape-based parameters for traditional 3D multimedia visual behavior can significantly reduce the stability of image simulation. In this paper, the video feature enhancement method of wavelet and human eye vision is used to set the initialization parameters of the ant colony method calculation and calculate the video pixel to different food source distances as a way to obtain the amount of information of each path. The interface pixel affiliation is calculated to make a judgment on whether the amount of information is greater than the initial threshold. The contrast between the detail components of the interface at different scales is enhanced using the wavelet multistage method, and the video visual feedback enhancement is achieved by extracting the interface edge regions in each scale through the local shareholding method. Then the multimedia video representation of human behavior as silhouette sequence, extract the human behavior motion trajectory as large-scale multimedia visual behavior features, and extract the edge features of multimedia vision, and express the results as visual detail features in multimedia video. Finally, a multimedia visual human behavior state model is formed, and the patio-temporal sequence of the model is estimated by fusing it with particle filtering approximation inference to complete the behavioral simulation analysis of multimedia vision. The experimental results show that the average recognition rate of the method in this paper is as high as 98%, the accuracy of behavioral feature extraction is as high as 96%, and the error rate of modeling is 0.02%.</p>
</abstract>

The reliance on shape-based parameters for traditional 3D multimedia visual behavior can significantly reduce the stability of image simulation. In this paper, the video feature enhancement method of wavelet and human eye vision is used to set the initialization parameters of the ant colony method calculation and calculate the video pixel to different food source distances as a way to obtain the amount of information of each path. The interface pixel affiliation is calculated to make a judgment on whether the amount of information is greater than the initial threshold. The contrast between the detail components of the interface at different scales is enhanced using the wavelet multistage method, and the video visual feedback enhancement is achieved by extracting the interface edge regions in each scale through the local shareholding method. Then the multimedia video representation of human behavior as silhouette sequence, extract the human behavior motion trajectory as large-scale multimedia visual behavior features, and extract the edge features of multimedia vision, and express the results as visual detail features in multimedia video. Finally, a multimedia visual human behavior state model is formed, and the patio-temporal sequence of the model is estimated by fusing it with particle filtering approximation inference to complete the behavioral simulation analysis of multimedia vision. The experimental results show that the average recognition rate of the method in this paper is as high as 98%, the accuracy of behavioral feature extraction is as high as 96%, and the error rate of modeling is 0.02%.

Títul Mobile visual media design based on digital media technology and augmented reality

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

The reliance on shape-based parameters for traditional 3D multimedia visual behavior can significantly reduce the stability of image simulation. In this paper,...

Títul Mobile visual media design based on digital media technology and augmented reality

Online veröffentlicht: 02. Juni 2023

Seitenbereich: -

Eingereicht: 19. Juni 2022

Akzeptiert: 21. Okt. 2022

DOI: https://doi.org/10.2478/amns.2023.1.00184

Schlüsselwörter
3D multimedia, augmented reality, initial shareholding, wavelet multistage method, visual feedback

© 2023 Baoquan Luo, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Títul Mobile visual media design based on digital media technology and augmented reality

Online veröffentlicht: 02. Juni 2023

Seitenbereich: -

Eingereicht: 19. Juni 2022

Akzeptiert: 21. Okt. 2022

DOI: https://doi.org/10.2478/amns.2023.1.00184

Schlüsselwörter3D multimedia, augmented reality, initial shareholding, wavelet multistage method, visual feedback

© 2023 Baoquan Luo, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Schlüsselwörter
3D multimedia, augmented reality, initial shareholding, wavelet multistage method, visual feedback