GroundSet: A Cadastral-Grounded Dataset for Spatial Understanding with Vector Data
This addresses a critical gap in Earth Observation for applications like urban planning and disaster management, though it is incremental as it focuses on dataset creation and validation.
The authors tackled the problem of poor fine-grained spatial understanding in multimodal large language models for remote sensing by introducing a large-scale dataset grounded in cadastral vector data, comprising 3.8 million annotated objects across 510k images, and showed that high-fidelity supervision enables standard architectures to master spatial grounding without complex modifications.
Precise spatial understanding in Earth Observation is essential for translating raw aerial imagery into actionable insights for critical applications like urban planning, environmental monitoring and disaster management. However, Multimodal Large Language Models exhibit critical deficiencies in fine-grained spatial understanding within Remote Sensing, primarily due to a reliance on limited or repurposed legacy datasets. To bridge this gap, we introduce a large-scale dataset grounded in verifiable cadastral vector data, comprising 3.8 million annotated objects across 510k high-resolution images with 135 granular semantic categories. We validate this resource through a comprehensive instruction-tuning benchmark spanning seven spatial reasoning tasks. Our evaluation establishes a robust baseline using a standard LLaVA architecture. We show that while current RS-specialized and commercial models (e.g., Gemini) struggle in zero-shot settings, high-fidelity supervision effectively bridges this gap, enabling standard architectures to master fine-grained spatial grounding without complex architectural modifications.