AerialVG: A Challenging Benchmark for Aerial Visual Grounding by Exploring Positional Relations

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liu, Junli, Chen, Qizhi, Wang, Zhigang, Tang, Yiwen, Zhang, Yiting, Yan, Chi, Wang, Dong, Li, Xuelong, Zhao, Bin
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908581621661696
author Liu, Junli
Chen, Qizhi
Wang, Zhigang
Tang, Yiwen
Zhang, Yiting
Yan, Chi
Wang, Dong
Li, Xuelong
Zhao, Bin
author_facet Liu, Junli
Chen, Qizhi
Wang, Zhigang
Tang, Yiwen
Zhang, Yiting
Yan, Chi
Wang, Dong
Li, Xuelong
Zhao, Bin
contents Visual grounding (VG) aims to localize target objects in an image based on natural language descriptions. In this paper, we propose AerialVG, a new task focusing on visual grounding from aerial views. Compared to traditional VG, AerialVG poses new challenges, \emph{e.g.}, appearance-based grounding is insufficient to distinguish among multiple visually similar objects, and positional relations should be emphasized. Besides, existing VG models struggle when applied to aerial imagery, where high-resolution images cause significant difficulties. To address these challenges, we introduce the first AerialVG dataset, consisting of 5K real-world aerial images, 50K manually annotated descriptions, and 103K objects. Particularly, each annotation in AerialVG dataset contains multiple target objects annotated with relative spatial relations, requiring models to perform comprehensive spatial reasoning. Furthermore, we propose an innovative model especially for the AerialVG task, where a Hierarchical Cross-Attention is devised to focus on target regions, and a Relation-Aware Grounding module is designed to infer positional relations. Experimental results validate the effectiveness of our dataset and method, highlighting the importance of spatial reasoning in aerial visual grounding. The code and dataset will be released.
format Preprint
id arxiv_https___arxiv_org_abs_2504_07836
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AerialVG: A Challenging Benchmark for Aerial Visual Grounding by Exploring Positional Relations
Liu, Junli
Chen, Qizhi
Wang, Zhigang
Tang, Yiwen
Zhang, Yiting
Yan, Chi
Wang, Dong
Li, Xuelong
Zhao, Bin
Computer Vision and Pattern Recognition
Artificial Intelligence
Visual grounding (VG) aims to localize target objects in an image based on natural language descriptions. In this paper, we propose AerialVG, a new task focusing on visual grounding from aerial views. Compared to traditional VG, AerialVG poses new challenges, \emph{e.g.}, appearance-based grounding is insufficient to distinguish among multiple visually similar objects, and positional relations should be emphasized. Besides, existing VG models struggle when applied to aerial imagery, where high-resolution images cause significant difficulties. To address these challenges, we introduce the first AerialVG dataset, consisting of 5K real-world aerial images, 50K manually annotated descriptions, and 103K objects. Particularly, each annotation in AerialVG dataset contains multiple target objects annotated with relative spatial relations, requiring models to perform comprehensive spatial reasoning. Furthermore, we propose an innovative model especially for the AerialVG task, where a Hierarchical Cross-Attention is devised to focus on target regions, and a Relation-Aware Grounding module is designed to infer positional relations. Experimental results validate the effectiveness of our dataset and method, highlighting the importance of spatial reasoning in aerial visual grounding. The code and dataset will be released.
title AerialVG: A Challenging Benchmark for Aerial Visual Grounding by Exploring Positional Relations
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2504.07836