EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Multimodal Large Language Models for Visual Attribute Inference in iRAP Road Attribute Coding

Horia Ameen, Natchapon Jongwiriyanurak, Jesús Balado, Mario Soilan

Infrastructures · 12.03.2026 · 11 (3) · 95

DOI: 10.3390/infrastructures11030095 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Infrastructures, online erschienen am 12.03.2026. Verfasst von Horia Ameen und 3 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 46 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministerio de Ciencia, Innovación y Universidades, Xunta de Galicia.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Road safety assessment is essential for reducing traffic fatalities, with road infrastructure contributing to a substantial proportion of crashes worldwide. International frameworks such as the International Road Assessment Program (iRAP) define standardized attributes for infrastructure auditing; however, many of these attributes remain challenging to automate using imagery alone. This study evaluates V-RoAst (visual question answering for road assessment), a public dataset of road images that are annotated with iRAP-style attributes, using state-of-the-art multimodal large language models (MLLMs), specifically Gemini 2.0 and Gemini 2.5. The analysis focuses on how prompt design influences the accuracy and stability of single image iRAP inference. A token-efficient reduced prompt is developed that preserves the iRAP schema while removing single-class constants, hard-coded administrative fields, and derived or non-visual codes, retaining only visually interpretable attributes. Performance is compared with the original full multi-attribute prompt and single attribute prompts using a fixed evaluation protocol incorporating majority voting, bootstrap 95% confidence intervals, and per-code sample-size checks. Results indicate only minor performance differences between Gemini 2.0 and Gemini 2.5, while prompt optimization produces the most consistent gains, improving macro-F1 scores and tightening confidence intervals for visually grounded attributes such as roadside severity, intersection channelization, and service-road presence. Token analysis shows an approximate 30% reduction in prompt length, reducing computational cost and truncation risk. Overall, the findings demonstrate that prompt scope has a greater impact than model version in image-only iRAP coding, offering practical guidance for scalable infrastructure assessment.

Förderung

  • Ministerio de Ciencia, Innovación y Universidades
  • Xunta de Galicia

Zitieren

APA

Ameen, H., Jongwiriyanurak, N., Balado, J., & Soilan, M. (2026). Multimodal Large Language Models for Visual Attribute Inference in iRAP Road Attribute Coding. Infrastructures, 11(3), 95. https://doi.org/10.3390/infrastructures11030095

Vancouver

Ameen H, Jongwiriyanurak N, Balado J, Soilan M. Multimodal Large Language Models for Visual Attribute Inference in iRAP Road Attribute Coding. Infrastructures. 2026;11(3):95. doi:10.3390/infrastructures11030095

BibTeX
@article{Ameen2026,
  author = {Ameen, Horia and Jongwiriyanurak, Natchapon and Balado, Jesús and Soilan, Mario},
  title = {Multimodal Large Language Models for Visual Attribute Inference in iRAP Road Attribute Coding},
  journal = {Infrastructures},
  year = {2026},
  volume = {11},
  number = {3},
  pages = {95},
  doi = {10.3390/infrastructures11030095}
}
Kostenlos starten →