Salvato in:
Dettagli Bibliografici
Autori principali: Dimopoulos, Alexander Nikitas, Grasso, Joseph, Beltz, John
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2604.23095
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913061728681984
author Dimopoulos, Alexander Nikitas
Grasso, Joseph
Beltz, John
author_facet Dimopoulos, Alexander Nikitas
Grasso, Joseph
Beltz, John
contents Indoor environments lack the spatial intelligence infrastructure that GPS provides outdoors; first responders arriving at unfamiliar buildings typically have no machine-readable map of safety equipment. Prior work on 3D semantic segmentation for public safety identified two barriers: scarcity of labeled indoor training data and poor recognition of small safety-critical features by native point-cloud methods. This paper presents INSIGHT, a zero-target-domain-annotation pipeline that projects 2D image understanding into 3D metric space via registered RGB-D data. Two interchangeable vision stacks share a common 3D back end: a SAM3 foundation-model stack for text-prompted segmentation, and a traditional CV stack (open-set detection, VQA, OCR) whose intermediate outputs are independently inspectable. Evaluated on all seven subareas of Stanford 2D-3D-S (70{,}496 images), the pipeline produces Pointcept-schema-compatible labeled point clouds and ISO~19164-compliant scene graphs with ${\sim}10^{4}{\times}$ compression; role-filtered payloads transmit in ${<}15$\,s at 1\,Mbps over FirstNet Band~14. We report per-point labeling accuracy on 7 shared classes, detection sensitivity for 15 safety-critical classes absent from public 3D benchmarks alongside code-capped deployable estimates, and inter-pipeline complementarity, demonstrating that 2D-to-3D semantic transfer addresses the labeled-data bottleneck while scene graphs provide building intelligence compact enough for field deployment.
format Preprint
id arxiv_https___arxiv_org_abs_2604_23095
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle INSIGHT: Indoor Scene Intelligence from Geometric-Semantic Hierarchy Transfer for Public~Safety
Dimopoulos, Alexander Nikitas
Grasso, Joseph
Beltz, John
Computer Vision and Pattern Recognition
Emerging Technologies
Indoor environments lack the spatial intelligence infrastructure that GPS provides outdoors; first responders arriving at unfamiliar buildings typically have no machine-readable map of safety equipment. Prior work on 3D semantic segmentation for public safety identified two barriers: scarcity of labeled indoor training data and poor recognition of small safety-critical features by native point-cloud methods. This paper presents INSIGHT, a zero-target-domain-annotation pipeline that projects 2D image understanding into 3D metric space via registered RGB-D data. Two interchangeable vision stacks share a common 3D back end: a SAM3 foundation-model stack for text-prompted segmentation, and a traditional CV stack (open-set detection, VQA, OCR) whose intermediate outputs are independently inspectable. Evaluated on all seven subareas of Stanford 2D-3D-S (70{,}496 images), the pipeline produces Pointcept-schema-compatible labeled point clouds and ISO~19164-compliant scene graphs with ${\sim}10^{4}{\times}$ compression; role-filtered payloads transmit in ${<}15$\,s at 1\,Mbps over FirstNet Band~14. We report per-point labeling accuracy on 7 shared classes, detection sensitivity for 15 safety-critical classes absent from public 3D benchmarks alongside code-capped deployable estimates, and inter-pipeline complementarity, demonstrating that 2D-to-3D semantic transfer addresses the labeled-data bottleneck while scene graphs provide building intelligence compact enough for field deployment.
title INSIGHT: Indoor Scene Intelligence from Geometric-Semantic Hierarchy Transfer for Public~Safety
topic Computer Vision and Pattern Recognition
Emerging Technologies
url https://arxiv.org/abs/2604.23095