IPDiff: Diffusion-driven ORSI Salient Object Detection with Information Reconstruction and Multi-Prior Guidance

2026-07-04

AI Summary

This work addresses the limitation of existing salient object detection methods for optical remote sensing images, which rely on static inference and cannot correct errors during testing. To overcome this, the study formulates the task as a conditional diffusion process for the first time and introduces a dynamic parameter optimization mechanism that iteratively refines the saliency map at inference time. The proposed method integrates an information reconstruction attention module with multi-prior guidance and employs joint supervision in both spatial and spectral domains during training. A hybrid loss function is designed to effectively drive the denoising network’s learning process. Extensive experiments demonstrate that the approach significantly outperforms 46 state-of-the-art methods on the ORSSD, EORSSD, and ORSI-4199 benchmarks, achieving new state-of-the-art performance.

Abstract

Existing Salient Object Detection in Optical Remote Sensing Image (ORSI-SOD) methods mainly adopt the static inference strategy, which uses fixed trained model parameters for saliency inference in the testing phase. This means that even if the generated saliency map has errors, it cannot be further optimized. In this paper, we propose the novel IPDiff, a Diffusion-driven ORSI-SOD method with Information Reconstruction and Multi-Prior Guidance. We build IPDiff based on a unique dynamic optimization strategy, which endows IPDiff with the ability to iteratively optimize saliency maps with a dynamic parameter. Specifically, we formulate ORSI-SOD as a conditional diffusion problem in IPDiff. IPDiff first extracts informative conditional priors from ORSIs, including the saliency prior and the hierarchical priors, in the prior network with the assistance of the information reconstruction-driven attention module. The saliency prior can provide positional information of salient objects, while the hierarchical priors can provide specific detail and semantic information of salient objects. Under the guidance of these priors, IPDiff then iteratively denoises random noise as the timestep dynamically changes in the denoising network, generating saliency maps that are close to ground truths. Notably, we simultaneously supervise IPDiff in both spatial and spectral domains through a hybrid loss function to achieve efficient network training. Comprehensive experiments on public ORSSD, EORSSD, and ORSI-4199 datasets demonstrate that our proposed IPDiff achieves the best performance compared to 46 state-of-the-art methods. The code and results of our method are available at https://github.com/MathLee/IPDiff.

Problem

• Salient Object Detection
• Optical Remote Sensing Image
• Static Inference
• Saliency Map Optimization
• Dynamic Optimization

Innovation

• Diffusion-driven
• Dynamic optimization
• Information reconstruction
• Multi-prior guidance
• Salient object detection