Building extraction from oblique photogrammetry point clouds based on <scp>PointNet</scp>++ with attention mechanism
Bibliographic record
Abstract
Abstract Unmanned aircraft vehicles (UAVs) capture oblique point clouds in outdoor scenes that contain considerable building information. Building features extracted from images are affected by the viewing point, illumination, occlusion, noise and image conditions, which make building features difficult to extract. Currently, ground elevation changes can provide powerful aids for the extraction, and point cloud data can precisely reflect this information. Thus, oblique photogrammetry point clouds have significant research implications. Traditional building extraction methods involve the filtering and sorting of raw data to separate buildings, which cause the point clouds to lose spatial information and reduce the building extraction accuracy. Therefore, we develop an intelligent building extraction method based on deep learning that incorporates an attention mechanism module into the Samling and PointNet operations within the set abstraction layer of the PointNet++ network. To assess the efficacy of our approach, we train and extract buildings from a dataset created using UAV oblique point clouds from five regions in the city of Bengbu, China. Impressive performance metrics are achieved, including 95.7% intersection over union, 96.5% accuracy, 96.5% precision, 98.7% recall and 97.8% F1 score. And with the addition of attention mechanism, the overall training accuracy of the model is improved by about 3%. This method showcases potential for advancing the accuracy and efficiency of digital urbanization construction projects.
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 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.000 | 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) | 0.002 | 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".