English

DiffuSAM: Diffusion Guided Zero-Shot Object Grounding for Remote Sensing Imagery

Computer Vision and Pattern Recognition 2026-04-21 v1 Machine Learning

Abstract

Diffusion models have emerged as powerful tools for a wide range of vision tasks, including text-guided image generation and editing. In this work, we explore their potential for object grounding in remote sensing imagery. We propose a hybrid pipeline that integrates diffusion-based localization cues with state-of-the-art segmentation models such as RemoteSAM and SAM3 to obtain more accurate bounding boxes. By leveraging the complementary strengths of generative diffusion models and foundational segmentation models, our approach enables robust and adaptive object localization across complex scenes. Experiments demonstrate that our pipeline significantly improves localization performance, achieving over a 14% increase in Acc@0.5 compared to existing state-of-the-art methods.

Keywords

Cite

@article{arxiv.2604.18201,
  title  = {DiffuSAM: Diffusion Guided Zero-Shot Object Grounding for Remote Sensing Imagery},
  author = {Geet Sethi and Panav Shah and Ashutosh Gandhe and Soumitra Darshan Nayak},
  journal= {arXiv preprint arXiv:2604.18201},
  year   = {2026}
}

Comments

Accepted at ICLR 2026 ML4RS Workshop

R2 v1 2026-07-01T12:18:16.964Z