Smart City Gnosys

Smart city article details

Title A Quality Evaluation Method Of Lightweight Realistic 3D Models Based On Building Information Modeling
ID_Doc 3928
Authors Chen J.; Luo Y.; Zhang H.
Year 2023
Published 2023 4th International Conference on Computer Vision, Image and Deep Learning, CVIDL 2023
DOI http://dx.doi.org/10.1109/CVIDL58838.2023.10165778
Abstract Building information models and realistic three-dimensional models, which are the main sources of fundamental data for geographic information systems, have attracted a lot of attention in academia and business as smart cities, twin cities, and infrastructure construction have gained increasing momentum in recent years. The architectural design industry will undergo a paradigm shift as a result of the integration of building information models with fully-textured and fully-element realistic 3D models, which will lower costs and increase construction efficiency. Nonetheless, due to inherent shortcomings of realistic 3D models, it is still difficult to introduce these models to building information models forward design platforms. In the present work, a method was proposed to evaluate the quality of the lightweight realistic 3D model outputs for application of the lightweight realistic 3D model in building information models forward design, and the model outputs were evaluated by two quality evaluation indicators- visual effect and positioning accuracy, in a stepped manner. The thresholds of the evaluation indicators were determined based on the idea of plotting scale in digital line graphs (DLGs), and the quality score of the 3D model was calculated by the preset weights of the quality evaluation indicators to assess the quality of the 3D model. Experiments revealed that the lightweight realistic 3D model quality evaluation method could minimize the volume of data of the 3D model without affecting the visual effect and positioning accuracy of the model. The problem that the realistic 3D model could not be effectively introduced to the building information model forward design platform due to its massive quantity of data could be solved by the proposed method, which could effectively control the impact of the lightweight processing on the quality of the realistic 3D model. © 2023 IEEE.
Author Keywords lightweight processing; multi-attribute comprehensive evaluation; realistic 3D model


Similar Articles


Id Similarity Authors Title Published
117 View0.85Liu W.; Zang Y.; Xiong Z.; Bian X.; Wen C.; Lu X.; Wang C.; Marcato J., Jr.; Gonçalves W.N.; Li J.3D Building Model Generation From Mls Point Cloud And 3D Mesh Using Multi-Source Data FusionInternational Journal of Applied Earth Observation and Geoinformation, 116 (2023)