Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Are VLMs Lost Between Sky and Space? LinkS$^2$Bench for UAV-Satellite Dynamic Cross-View Spatial Intelligence

Dian Liu, Jie Feng, Di Li, Yuhui Zheng, Guanbin Li, Weisheng Dong, Guangming Shi

Abstract

Synergistic spatial intelligence between UAVs and satellites is indispensable for emergency response and security operations, as it uniquely integrates macro-scale global coverage with dynamic, real-time local perception. However, the capacity of Vision-Language Models (VLMs) to master this complex interplay remains largely unexplored. This gap persists primarily because existing benchmarks are confined to isolated Unmanned Aerial Vehicle (UAV) videos or static satellite imagery, failing to evaluate the dynamic local-to-global spatial mapping essential for comprehensive cross-view reasoning. To bridge this gap, we introduce LinkS$^2$Bench, the first comprehensive benchmark designed to evaluate VLMs' wide-area, dynamic cross-view spatial intelligence. LinkS$^2$Bench links 1,022 minutes of dynamic UAV footage with high-resolution satellite imagery covering over 200 km$^2$. Through an LMM-assisted pipeline and rigorous human annotation, we constructed 17.9k high-quality question-answer pairs comprising 12 fine-grained tasks across four dimensions: perception, localization, relation, and reasoning. Evaluations of 18 representative VLMs reveal a substantial gap compared to human baselines, identifying accurate cross-view dynamic alignment as the critical bottleneck. To alleviate this, we design a Cross-View Alignment Adapter, demonstrating that explicit alignment significantly improves model performance. Furthermore, fine-tuning experiments underscore the potential of LinkS$^2$Bench in advancing VLM adaptation for complex spatial reasoning.

Are VLMs Lost Between Sky and Space? LinkS$^2$Bench for UAV-Satellite Dynamic Cross-View Spatial Intelligence

Abstract

Synergistic spatial intelligence between UAVs and satellites is indispensable for emergency response and security operations, as it uniquely integrates macro-scale global coverage with dynamic, real-time local perception. However, the capacity of Vision-Language Models (VLMs) to master this complex interplay remains largely unexplored. This gap persists primarily because existing benchmarks are confined to isolated Unmanned Aerial Vehicle (UAV) videos or static satellite imagery, failing to evaluate the dynamic local-to-global spatial mapping essential for comprehensive cross-view reasoning. To bridge this gap, we introduce LinkSBench, the first comprehensive benchmark designed to evaluate VLMs' wide-area, dynamic cross-view spatial intelligence. LinkSBench links 1,022 minutes of dynamic UAV footage with high-resolution satellite imagery covering over 200 km. Through an LMM-assisted pipeline and rigorous human annotation, we constructed 17.9k high-quality question-answer pairs comprising 12 fine-grained tasks across four dimensions: perception, localization, relation, and reasoning. Evaluations of 18 representative VLMs reveal a substantial gap compared to human baselines, identifying accurate cross-view dynamic alignment as the critical bottleneck. To alleviate this, we design a Cross-View Alignment Adapter, demonstrating that explicit alignment significantly improves model performance. Furthermore, fine-tuning experiments underscore the potential of LinkSBench in advancing VLM adaptation for complex spatial reasoning.

Paper Structure

This paper contains 20 sections, 5 equations, 6 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Spatial Intelligence of VLMs
  4. Benchmarks for Spatial Intelligence
  5. LinkS$^2$Bench
  6. Overview
  7. Task Design
  8. Benchmark Construction Pipeline
  9. Experiments
  10. Experimental Setup
  11. Benchmark Models
  12. Metric Design
  13. Human Level Performance.
  14. Model Comparison
  15. Correlation Analysis
  16. ...and 5 more sections

Figures (6)

  • Figure 1: Benchmark Statistics. (a) The distribution of tasks across four main categories. (b) Distribution of data sources across different video durations.
  • Figure 2: LinkS$^2$Bench comprises 12 fine-grained task types categorized into four primary evaluation dimensions: Perception, Location, Relation, and Reasoning. For visualization purposes, the satellite images are cropped to highlight the annotations; the actual benchmark samples preserve the full original spatial context.
  • Figure 3: LinkS$^2$Bench curation pipeline includes data collection, question generation, human annotation and quality control.
  • Figure 4: Pairwise task correlations based on VLM performance. Higher values suggest that VLMs perform more similarly on the two tasks, indicating a greater overlap in the abilities required to solve them.
  • Figure 5: Distribution of error types via manual inspection.
  • ...and 1 more figures