Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

ORSIFlow: Saliency-Guided Rectified Flow for Optical Remote Sensing Salient Object Detection

Haojing Chen, Yutong Li, Zhihang Liu, Tao Tan, Haoyu Bian, Qiuju Ma

Abstract

Optical Remote Sensing Image Salient Object Detection (ORSI-SOD) remains challenging due to complex backgrounds, low contrast, irregular object shapes, and large variations in object scale. Existing discriminative methods directly regress saliency maps, while recent diffusion-based generative approaches suffer from stochastic sampling and high computational cost. In this paper, we propose ORSIFlow, a saliency-guided rectified flow framework that reformulates ORSI-SOD as a deterministic latent flow generation problem. ORSIFlow performs saliency mask generation in a compact latent space constructed by a frozen variational autoencoder, enabling efficient inference with only a few steps. To enhance saliency awareness, we design a Salient Feature Discriminator for global semantic discrimination and a Salient Feature Calibrator for precise boundary refinement. Extensive experiments on multiple public benchmarks show that ORSIFlow achieves state-of-the-art performance with significantly improved efficiency. Codes are available at: https://github.com/Ch3nSir/ORSIFlow.

ORSIFlow: Saliency-Guided Rectified Flow for Optical Remote Sensing Salient Object Detection

Abstract

Optical Remote Sensing Image Salient Object Detection (ORSI-SOD) remains challenging due to complex backgrounds, low contrast, irregular object shapes, and large variations in object scale. Existing discriminative methods directly regress saliency maps, while recent diffusion-based generative approaches suffer from stochastic sampling and high computational cost. In this paper, we propose ORSIFlow, a saliency-guided rectified flow framework that reformulates ORSI-SOD as a deterministic latent flow generation problem. ORSIFlow performs saliency mask generation in a compact latent space constructed by a frozen variational autoencoder, enabling efficient inference with only a few steps. To enhance saliency awareness, we design a Salient Feature Discriminator for global semantic discrimination and a Salient Feature Calibrator for precise boundary refinement. Extensive experiments on multiple public benchmarks show that ORSIFlow achieves state-of-the-art performance with significantly improved efficiency. Codes are available at: https://github.com/Ch3nSir/ORSIFlow.

Paper Structure

This paper contains 12 sections, 15 equations, 4 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Methodlogy
  4. Overall Architecture
  5. Salient Feature Conditional Network
  6. Latent Space Flow Network
  7. Eexperiment
  8. Experimental Setup
  9. Performance Comparison
  10. Ablation Studies
  11. Efficiency Analysis
  12. Conclusion

Figures (4)

  • Figure 1: The left panel shows performance in terms of $S_{\alpha}$ on three datasets, while the right panel compares our method against a previous state-of-the-art approach. Red boxes indicate segmentation errors.
  • Figure 2: Overall architecture of the proposed method. The SFCN extracts multi-scale conditional features via SFD and SFC modules , while the LSFN learns a deterministic velocity field to transform the Gaussian distribution into saliency maps within the VAE latent space.
  • Figure 3: Qualitative comparison of ORSIFlow and other state-of-the-art methods
  • Figure 4: Visual results of the effectiveness of our modules.