MétaCan
Menu
Back to cohort

Box-Supervised Instance Segmentation For Building Rooftop Delineation From High Spatial Resolution Remotely Sensed Imagery

2025· article· W4416727601 on OpenAlexaff
Hongjie He, Nan Chen, Lingfei Ma, Jonathan Li

Bibliographic record

Venuenot available
Typearticle
Language
FieldComputer Science
TopicAdvanced Neural Network Applications
Canadian institutionsUniversity of Waterloo
FundersChina Scholarship CouncilNational Natural Science Foundation of China
KeywordsSegmentationScale (ratio)Image segmentationBounding overwatchBuilding modelMinimum bounding boxImage resolutionFeature (linguistics)

Abstract

fetched live from OpenAlex

In this work, we propose Box2Boundary for box-supervised instance segmentation in building rooftop delineation. The proposed Box2Boundary is developed based on the state-of-the-art(SOTA) bounding box-supervised instance segmentation method Box2Mask. We incorporate InternImage as the backbone for better feature extraction and employ Dynamic Scale Training (DST) during model training. InternImage and DST are integrated into Box2Boundary to address scale variance issues and the complexity of building boundaries in deep learning-based methods. Experiments on the WuHan University building dataset (WHU) demonstrate SOTA performance in box-supervised building rooftop delineation, with an Average Precision (AP) value of 48.7%, surpassing the previous SOTA method Box2Mask by 2%. While it achieves poor performance in building damage assessment using the xBD dataset, we provide a preliminary test of box-supervised instance segmentation for this task. Thus, the proposed Box2Boundary is validated as a practical method for building rooftop delineation, especially when high-quality building annotations are unavailable.

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 imitation

Not 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.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.002
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.005
Threshold uncertainty score0.011

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0010.002
Meta-epidemiology (narrow)0.0020.001
Meta-epidemiology (broad)0.0010.002
Bibliometrics0.0020.001
Science and technology studies0.0000.001
Scholarly communication0.0010.002
Open science0.0020.001
Research integrity0.0010.001
Insufficient payload (model declined to judge)0.0030.002

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.

Opus teacher head0.022
GPT teacher head0.291
Teacher spread0.269 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreMethods

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".

Quick stats

Citations0
Published2025
Admission routes1
Has abstractyes

Explore more

Same topicAdvanced Neural Network ApplicationsFrench-language works237,207