Remote sensing models improve locating changes in satellite images
From Coordinates to Candidate Regions: Temporal Change Localization via Region Selection in Remote Sensing Multimodal LLMs
Computer Vision and Pattern RecognitionComputation and Language
Summary
Locating specific changes or objects in satellite images is difficult because these targets are often small and close together. The authors propose a new method where the model selects from suggested regions instead of predicting exact box coordinates. This makes it easier for the model to find and describe changes over time in multiple images. Their approach works better than older methods at spotting changes and grounding visual details, while still understanding the scenes well. They also analyze which parts of their system contribute most to the improvements.
remote sensingmultimodal large language modelssatellite imagerytemporal change localizationregion selectionvisual groundingbounding box coordinatesregion proposalvisual question answering
Authors
Juwan Chung, Sungjune Park, Yeongyun Kim, Yong Man Ro
Abstract
Remote sensing multimodal large language models (RS-MLLMs) have advanced scene understanding and visual question answering over satellite imagery, yet localizing specific objects or changed regions remains challenging. Existing approaches rely on generating bounding box coordinates as token sequences, which is fragile for the small, densely packed objects common in remote sensing and increasingly error-prone when multiple targets must be localized simultaneously. In this work, we present an RS-specific formulation of the region selection paradigm, previously explored in natural-image MLLMs, and extend it to temporal change localization over multi-image sequences. Our framework employs a text-conditioned region proposal module, encodes each candidate as special tokens carrying per-frame visual features enriched with spatial and temporal cues, and lets the LLM localize targets by selecting region tokens in its response. We construct a multi-task training and evaluation suite spanning localization, referring expression, visual grounding, and understanding tasks across single-image and multi-temporal settings. Experiments show that our approach substantially outperforms coordinate-generation baselines on temporal change localization, while improving single-image visual grounding and maintaining competitive understanding performance. Oracle analysis decomposes the contributions of the region proposer and the LLM selector, providing diagnostic insight unique to this framework. Our code will be available at https://github.com/juwan-kr/RS-RegionSelect.