Peter Kán, Andrija Kurtic, Mohamed Radwan, Jorge M. Loáiciga Rodríguez
Automatic Interior Design in Augmented Reality Based on Hierarchical Tree of Procedural Rules
Electronics, 10(3):1-17, 2021.

Information

  • Publication Type: Journal Paper (without talk)
  • Workgroup(s)/Project(s): not specified
  • Date: 2021
  • DOI: 10.3390/electronics10030245
  • Journal: Electronics
  • Number: 3
  • Open Access: yes
  • Pages (from): 1
  • Pages (to): 17
  • Volume: 10
  • Keywords: interior design, augmented reality, 3D content generation, user study, personalized recommender

Abstract

Augmented reality has a high potential in interior design due to its capability of visualizing numerous prospective designs directly in a target room. In this paper, we present our research on utilization of augmented reality for interactive and personalized furnishing. We propose a new algorithm for automated interior design which generates sensible and personalized furniture configurations. This algorithm is combined with mobile augmented reality system to provide a user with an interactive interior design try-out tool. Personalized design is achieved via a recommender system which uses user preferences and room data as input. We conducted three user studies to explore different aspects of our research. The first study investigated the user preference between augmented reality and on-screen visualization for interactive interior design. In the second user study, we studied the user preference between our algorithm for automated interior design and optimization-based algorithm. Finally, the third study evaluated the probability of sensible design generation by the compared algorithms. The main outcome of our research suggests that augmented reality is viable technology for interactive home furnishing.

Additional Files and Images

Additional images and videos

Additional files

Weblinks

BibTeX

@article{Kan_Peter-2021-MDPI,
  title =      "Automatic Interior Design in Augmented Reality Based on
               Hierarchical Tree of Procedural Rules",
  author =     "Peter K\'{a}n and Andrija Kurtic and Mohamed Radwan and
               Jorge M. Lo\'{a}iciga Rodr\'{i}guez",
  year =       "2021",
  abstract =   "Augmented reality has a high potential in interior design
               due to its capability of visualizing numerous prospective
               designs directly in a target room. In this paper, we present
               our research on utilization of augmented reality for
               interactive and personalized furnishing. We propose a new
               algorithm for automated interior design which generates
               sensible and personalized furniture configurations. This
               algorithm is combined with mobile augmented reality system
               to provide a user with an interactive interior design
               try-out tool. Personalized design is achieved via a
               recommender system which uses user preferences and room data
               as input. We conducted three user studies to explore
               different aspects of our research. The first study
               investigated the user preference between augmented reality
               and on-screen visualization for interactive interior design.
               In the second user study, we studied the user preference
               between our algorithm for automated interior design and
               optimization-based algorithm. Finally, the third study
               evaluated the probability of sensible design generation by
               the compared algorithms. The main outcome of our research
               suggests that augmented reality is viable technology for
               interactive home furnishing.",
  doi =        "10.3390/electronics10030245",
  journal =    "Electronics",
  number =     "3",
  volume =     "10",
  pages =      "1--17",
  keywords =   "interior design, augmented reality, 3D content generation,
               user study, personalized recommender",
  URL =        "https://www.cg.tuwien.ac.at/research/publications/2021/Kan_Peter-2021-MDPI/",
}