Harnessing machine learning for rapid and cost-efficient 3D geometry generation in neighborhood energy modeling
Bibliographic record
Abstract
Accurate and scalable 3D building stock modeling is essential for reliable energy modeling, urban planning, and the design of sustainable, net-zero communities. This study introduces an automated method that generates 3D building models with minimal computational resources using OpenStreetMap footprints and street view images. The approach integrates image processing with machine learning techniques, including convolutional neural networks (CNN), Faster R-CNN, and semantic segmentation networks (SSN), from limited input data, one footprint and four façade images per building. Validation against manual measurements demonstrates an average volume accuracy of 95%, while modeling times are reduced from 3,600–19,800 s (manual process) to 458–466 s. Furthermore, energy simulations based on these models show reasonable agreement with energy audits, with variations ranging from 0.6% to 8.9%. The novelty of this work lies in its ability to combine open data sources, discrete methods and ML-based image processing for rapid, cost-effective, and interoperable 3D stock modeling. This framework accelerates large-scale digital twin development and supports integration with open BIM standards for construction and facility management. • ML techniques streamline urban digital twin creation for sustainable cities. • Method focused on low computational cost and modeling time reduction. • Automated geometry modeling using ML for building stock energy modeling. • Adaptable method supports various detail levels in geometric data collection. • Integration of various imaging datasets for detailed energy modeling.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".