Scholarlink

<p>The research aims to address the shortcomings of existing interior design generative models on design accuracy, diversity, and satisfaction of user personalized needs. Therefore, the research proposes an intelligent generation and optimization system based on Generative Adversarial Networks (GANs). The system generates fast and high-quality interior design solutions through adversarial training of generators and discriminators, combined with virtual reality technology. The research methods include building adversarial networks, optimizing design schemes through the competition between generators and discriminators, and providing immersive experiences through VR technology to enhance user interaction. The proposed model demonstrates superiority in key performance indicators such as image generation speed (32.41&#xa0;s/frame), resolution (2,048 × 2,048 pixels), color matching accuracy (94.12%), spatial function matching (88.61%), design diversity (95.45%), and visual effects (92.78%). Meanwhile, the model also performs well in feature point matching accuracy and running time, with the highest matching accuracy and shortest running time among all feature point quantities. The research results indicate that the proposed model can significantly improves the generation effectiveness and standard of interior design schemes, meet the personalized needs of users, and offers significant potential for advancing the automation and intelligence progress of interior design.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Intelligent generation and optimization system for interior design scheme based on GANs generation and VR immersive experience

  • 期刊论文

  • 发表日期：

    2025年8月13日

查看全文
  • Liuyu Zhang,
  • Quan Gan

摘要

The research aims to address the shortcomings of existing interior design generative models on design accuracy, diversity, and satisfaction of user personalized needs. Therefore, the research proposes an intelligent generation and optimization system based on Generative Adversarial Networks (GANs). The system generates fast and high-quality interior design solutions through adversarial training of generators and discriminators, combined with virtual reality technology. The research methods include building adversarial networks, optimizing design schemes through the competition between generators and discriminators, and providing immersive experiences through VR technology to enhance user interaction. The proposed model demonstrates superiority in key performance indicators such as image generation speed (32.41 s/frame), resolution (2,048 × 2,048 pixels), color matching accuracy (94.12%), spatial function matching (88.61%), design diversity (95.45%), and visual effects (92.78%). Meanwhile, the model also performs well in feature point matching accuracy and running time, with the highest matching accuracy and shortest running time among all feature point quantities. The research results indicate that the proposed model can significantly improves the generation effectiveness and standard of interior design schemes, meet the personalized needs of users, and offers significant potential for advancing the automation and intelligence progress of interior design.