A crowd-AI dynamic neural network hyperparameter optimization approach for image-driven social sensing applications

Zhang, Yang; Zong, Ruohan; Shang, Lanyu; Wang, Dong

doi:10.1016/j.knosys.2023.110864

“…Unlike static image processing tasks, animation scene design faces greater dynamism and variability. Zhang et al [15] explored the dynamic hyperparameter optimization problem in IASD applications, aiming to dynamically determine the optimal hyperparameter configuration for artificial intelligence-based IASD solutions. These configurations are manually set by animators or AI experts and are difficult to adapt to the complex dynamic changes in animation scene design.…”

Section: Related Workmentioning

confidence: 99%

Interactive Landscape Design Based on the Integration of Computer-Aided Design and Virtual Reality

Yang,

Huang

2024

CAD&A

0

View full text Add to dashboard Cite

Aiming at the problems of low rendering efficiency and the long time-consuming of traditional dynamic scene rendering methods, this study proposes an optimization method of animation design CAD and virtual reality dynamic scene rendering based on a transformer model. Through the construction of the animation design CAD model, the transformer model is used for animation design. Combined with spatial data conversion and graphics, the geometric transformation of animation design CAD model is carried out to realize the geometric transformation of graphics. By using primitive transformation, normal transformation, and lighting models combined with the data characteristics of dynamic scene rendering, the virtual reality dynamic scene rendering optimization model is constructed. Experimental results show that, compared with traditional methods, the proposed method can shorten the rendering time of animation design CAD and virtual reality dynamic scenes and effectively improve the rendering efficiency of animation design CAD and virtual reality dynamic scenes.

show abstract

Application of a Knowledge Based System in the Real Estate Market

Nagy

2023

2023 IEEE 23rd International Symposium on Computational Intelligence and Informatics (CINTI)

0

View full text Add to dashboard Cite

Virtual Reality-Based Interactive Experience of Music Creation Using a Diffusion Model

Chen,

Huang

2024

CAD&A

0

View full text Add to dashboard Cite

Aiming at the problems of low rendering efficiency and the long time-consuming of traditional dynamic scene rendering methods, this study proposes an optimization method of animation design CAD and virtual reality dynamic scene rendering based on a transformer model. Through the construction of the animation design CAD model, the transformer model is used for animation design. Combined with spatial data conversion and graphics, the geometric transformation of animation design CAD model is carried out to realize the geometric transformation of graphics. By using primitive transformation, normal transformation, and lighting models combined with the data characteristics of dynamic scene rendering, the virtual reality dynamic scene rendering optimization model is constructed. Experimental results show that, compared with traditional methods, the proposed method can shorten the rendering time of animation design CAD and virtual reality dynamic scenes and effectively improve the rendering efficiency of animation design CAD and virtual reality dynamic scenes.

show abstract

A crowd-AI dynamic neural network hyperparameter optimization approach for image-driven social sensing applications

Cited by 4 publications

References 38 publications

Interactive Landscape Design Based on the Integration of Computer-Aided Design and Virtual Reality

Interactive Landscape Design Based on the Integration of Computer-Aided Design and Virtual Reality

Application of a Knowledge Based System in the Real Estate Market

Virtual Reality-Based Interactive Experience of Music Creation Using a Diffusion Model

Contact Info

Product

Resources

About