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

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Haojing, Liu, Zhihang, Li, Yutong, Tan, Tao, Bian, Haoyu, Ma, Qiuju
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910161515315200
author Chen, Haojing
Liu, Zhihang
Li, Yutong
Tan, Tao
Bian, Haoyu
Ma, Qiuju
author_facet Chen, Haojing
Liu, Zhihang
Li, Yutong
Tan, Tao
Bian, Haoyu
Ma, Qiuju
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2603_28584
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle ORSIFlow: Saliency-Guided Rectified Flow for Optical Remote Sensing Salient Object Detection
Chen, Haojing
Liu, Zhihang
Li, Yutong
Tan, Tao
Bian, Haoyu
Ma, Qiuju
Computer Vision and Pattern Recognition
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.
title ORSIFlow: Saliency-Guided Rectified Flow for Optical Remote Sensing Salient Object Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.28584