EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Unified UAV Open-Vocabulary Semantic Segmentation: Benchmark Construction and LLM-Guided Text–Visual Enhancement

Kun Wang, Wei Li, Xiaopeng Liu, Duping Huang, Jiale Yang

Remote Sensing · 20.07.2026 · 18 (14) · 2408

DOI: 10.3390/rs18142408 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 20.07.2026. Verfasst von Kun Wang und 4 weiteren Autor:innen. Die Arbeit nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Open-vocabulary semantic segmentation (OVSS) of unmanned aerial vehicle (UAV) imagery aims to recognize arbitrary text-specified categories in aerial scenes, but existing OVSS models often suffer from UAV-domain shifts. To provide a reproducible testbed, this paper constructs a unified UAV OVSS benchmark by reorganizing multiple UAV segmentation datasets into cross-dataset transfer settings with explicit category harmonization and seen/unseen vocabulary analysis. Based on this benchmark, we propose UAV-OVSeg, a Cost Aggregation-style dense matching framework enhanced in two complementary directions: an LLM-guided Category Expansion Module that converts raw category names into structured UAV-aware descriptions, and a DINO-enhanced Geometric Feature Fusion Module that injects local structure into dense visual–text matching. Under SynDrone training, UAV-OVSeg achieves 65.5% mean intersection over union (mIoU) and 78.0% mean accuracy (mACC), improving upon the CAT-Seg baseline by 2.6 mIoU and 3.2 mACC. Under Aeroscapes training, it achieves 64.5% mIoU and 76.8% mACC, improving upon CAT-Seg by 2.6 mIoU and 3.1 mACC. Additional analyses of LLM variants, object scales, prompt sensitivity, boundary quality, and computational cost further verify the effectiveness and reproducibility of the proposed UAV-oriented text–visual enhancement strategy.

Zitieren

APA

Wang, K., Li, W., Liu, X., Huang, D., & Yang, J. (2026). Unified UAV Open-Vocabulary Semantic Segmentation: Benchmark Construction and LLM-Guided Text–Visual Enhancement. Remote Sensing, 18(14), 2408. https://doi.org/10.3390/rs18142408

Vancouver

Wang K, Li W, Liu X, Huang D, Yang J. Unified UAV Open-Vocabulary Semantic Segmentation: Benchmark Construction and LLM-Guided Text–Visual Enhancement. Remote Sensing. 2026;18(14):2408. doi:10.3390/rs18142408

BibTeX
@article{Wang2026,
  author = {Wang, Kun and Li, Wei and Liu, Xiaopeng and Huang, Duping and Yang, Jiale},
  title = {Unified UAV Open-Vocabulary Semantic Segmentation: Benchmark Construction and LLM-Guided Text–Visual Enhancement},
  journal = {Remote Sensing},
  year = {2026},
  volume = {18},
  number = {14},
  pages = {2408},
  doi = {10.3390/rs18142408}
}
Kostenlos starten →