A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Digital Automatic Modeling Algorithms for Building BIM Based on Convolutional Neural Networks
The complexity and diversity of building structures make traditional modeling methods difficult to adapt to various architectural forms and design styles. By introducing Inception models, the accuracy of BIM (Building Information Modeling) digital automatic modeling algorithms can be improved to ensure that the generated building models meet the requirements of actual design. A large amount of BIM data containing various building structures and design styles was collected, and the collected BIM data was preprocessed. The Inception model was constructed, including 1×1, 3×3, 5×5 convolution and global max pooling operations, and the cross entropy loss function was used to measure the performance of the model. The findings of the experiment demonstrated that the Inception model's average recognition accuracy for architectural features was 98.1%. Building element identification accuracy can be effectively increased with the use of Inception models, leading to more precise and effective digital automatic modeling of buildings.
Digital Automatic Modeling Algorithms for Building BIM Based on Convolutional Neural Networks
The complexity and diversity of building structures make traditional modeling methods difficult to adapt to various architectural forms and design styles. By introducing Inception models, the accuracy of BIM (Building Information Modeling) digital automatic modeling algorithms can be improved to ensure that the generated building models meet the requirements of actual design. A large amount of BIM data containing various building structures and design styles was collected, and the collected BIM data was preprocessed. The Inception model was constructed, including 1×1, 3×3, 5×5 convolution and global max pooling operations, and the cross entropy loss function was used to measure the performance of the model. The findings of the experiment demonstrated that the Inception model's average recognition accuracy for architectural features was 98.1%. Building element identification accuracy can be effectively increased with the use of Inception models, leading to more precise and effective digital automatic modeling of buildings.
Digital Automatic Modeling Algorithms for Building BIM Based on Convolutional Neural Networks
Ning, Jianwei (author)
2024-04-26
243248 byte
Conference paper
Electronic Resource
English
Convolutional Neural Networks and Regression Algorithms Supporting Buildings Facility Management
| DOAJ | 2023
Automatic soil classification method from CPTU data based on convolutional neural networks
| Springer Verlag | 2024
Estimating energy consumption for a building using dilated convolutional neural networks
European Patent Office | 2020
Robust Building Identification from Street Views Using Deep Convolutional Neural Networks
| DOAJ | 2024