A method for automatic identification of openings in buildings facades based on mobile LiDAR point clouds for assessing impacts of floodings
Bibliographic record
Abstract
Given the high frequency and major impact of flood events, decision-makers are in urgent need to have tools allowing them to predict or assess the impact of flood events on the population, such as flood simulation through digital 3D cities. While building’s lowest openings are more subject to potential damages during the flood, 3D building models of rural environments still lack this information. Thus, it would be required to provide the location of the building lowest openings, in the context of flood risk assessment. Unlike frequently developed methods in opening detection domain that benefit from the repetitive structure and symmetrical characterization of the openings on the facades of modern buildings in urban areas, this paper proposed a comprehensive approach that investigates low-rise residential houses of rural areas where openings have various shapes, sizes, and non-symmetrical positions on the facade. First, it proposed a generalized segmentation approach in a context involving various and complex facade structures, in the presence of frequent occlusion and noticeable point density changes. Second, it proposed a simple and consistent hole-based opening detection by effective elimination of window crossbars and curtains. Finally, by proposing an inventory of frequent challenges related to facade extraction and opening detection tasks, it enables a better understanding of the difficulties to help in providing more efficient and relevant solutions in rural residential contexts. Qualitative and quantitative evaluations were performed using an MLS real-world dataset of the Quebec Province, Canada. Related statistics revealed that the proposed approach could obtain good performance rates despite the complexity of the dataset, representative of the data acquired in real situations. Challenges regarding the characteristics of the MLS point cloud and the presence of large surrounding occlusions should be further investigated for obtaining more accurate opening information on the facade.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.004 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".