RS-RIE-Bench: Benchmarking Reasoning-Guided Remote Sensing Image Editing

2026-07-22

AI Summary

Existing image editing benchmarks struggle to evaluate the reasoning capabilities essential for remote sensing image editing, such as adherence to geographic rules, region-specific control, and sensor consistency. This work proposes the first reasoning-guided benchmark tailored for remote sensing image editing, encompassing temporal, causal, and spatial reasoning tasks, along with a multidimensional evaluation protocol that assesses target-region plausibility, non-target region preservation, and image quality consistency. For the first time, we introduce an automatic scoring mechanism based on multimodal large language models, complemented by hierarchical expert review, to systematically evaluate eight state-of-the-art models. Experimental results reveal that even the best-performing model achieves only 24.28% accuracy under stringent joint criteria, highlighting significant deficiencies in current approaches regarding causal and spatial reasoning.

Abstract

Remote sensing image editing aims to modify remote sensing images according to natural language instructions while preserving geographic rules and sensor observation characteristics. Existing benchmarks mainly target natural images or general visual scenes, and thus may not fully capture the reasoning, regional control, and sensor-consistency abilities required in remote sensing editing. To fill this gap, we introduce RS-RIE-Bench, the first benchmark for reasoning-guided remote sensing image editing. RS-RIE-Bench organizes tasks into three categories: temporal reasoning, causal reasoning, and spatial reasoning. These categories capture temporal evolution, causal consequence, and spatial imaging consistency in remote sensing scenes. The evaluation protocol covers three dimensions: target region plausibility, non-target region preservation, and image quality consistency. We further demonstrate the feasibility of MLLM-based evaluation through cross-judge consistency analysis and stratified expert review. Systematic evaluation on eight open-source and closed-source image editing models shows that current models still have clear limitations in reasoning-guided remote sensing editing. Even the strongest model achieves only 24.28\% overall accuracy under the strict joint-satisfaction criterion, while the mean relaxed joint-4 success rate across all eight models is 32.23\%. Causal reasoning and spatial reasoning remain especially challenging, and several open-source models are close to zero in some categories. These results show that RS-RIE-Bench can effectively reveal the limitations of current models in geographic reasoning, regional control, and sensor-consistent generation. It also provides a standardized benchmark and a clear research direction for future remote sensing intelligent editing models.

Problem

• remote sensing image editing
• reasoning
• benchmark
• sensor consistency
• regional control

Innovation

• reasoning-guided editing
• remote sensing image editing
• spatial-temporal reasoning
• sensor-consistent generation
• MLLM-based evaluation