EUVIMEDEUVIMED entdecken
Crossref

Learning Better UAV-Based Cross-View Object Geo-Localization from Multi-Modal Prompts: MoP-UAV Benchmark and MoPT Framework

Xiaohan Zhang, Zhangkai Shen, Si-Yuan Cao, Xiaokai Bai, Yiming Li, Zheheng Han, Zhe Wu, Qi Ming, Hui-Liang Shen

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (15) · 12843-12851

DOI: 10.1609/aaai.v40i15.38282 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Xiaohan Zhang und 8 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We present MoP-UAV, a new benchmark for UAV-based cross-view object geo-localization guided by multi-modal prompts. MoP-UAV supports fine-grained object-level cross-view localization under diverse prompt modalities, including natural language, bounding boxes, and click points. It offers potential for incorporating large foundation models like large language models (LLMs) and promotes the building of more flexible and intelligent UAV agents. Based on the benchmark, we propose MoPT, a multi-modal-prompt-guided tansformer that embeds prompts as token sequences and extract object location from UAV and satellite features via cross-attention. To enhance semantic consistency and performance, we further adopt a cross-view contrastive loss and propose a RefCOCOg-based pre-training strategy. Extensive experiments show that MoPT achieves robust localization under arbitrary prompt combinations. Notably, multi-modal-prompt training significantly boosts unimodal-prompt inference performance, highlighting the generalization benefits of multi-modal learning. MoPT trained with multi-modal prompts outperforms prior unimodal prompt works under the same setting.

Zitieren

APA

Zhang, X., Shen, Z., Cao, S., Bai, X., Li, Y., Han, Z., Wu, Z., Ming, Q., & Shen, H. (2026). Learning Better UAV-Based Cross-View Object Geo-Localization from Multi-Modal Prompts: MoP-UAV Benchmark and MoPT Framework. Proceedings of the AAAI Conference on Artificial Intelligence, 40(15), 12843-12851. https://doi.org/10.1609/aaai.v40i15.38282

Vancouver

Zhang X, Shen Z, Cao S, Bai X, Li Y, Han Z, et al. Learning Better UAV-Based Cross-View Object Geo-Localization from Multi-Modal Prompts: MoP-UAV Benchmark and MoPT Framework. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(15):12843-12851. doi:10.1609/aaai.v40i15.38282

BibTeX
@article{Zhang2026,
  author = {Zhang, Xiaohan and Shen, Zhangkai and Cao, Si-Yuan and Bai, Xiaokai and Li, Yiming and Han, Zheheng and Wu, Zhe and Ming, Qi and Shen, Hui-Liang},
  title = {Learning Better UAV-Based Cross-View Object Geo-Localization from Multi-Modal Prompts: MoP-UAV Benchmark and MoPT Framework},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {15},
  pages = {12843--12851},
  doi = {10.1609/aaai.v40i15.38282}
}
Kostenlos starten →