[Submitted on 30 Mar 2026 (v1), last revised 22 Oct 2026 (this version, v5)]
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.
Submission history
From: Haojing Chen [view email]
[v1] Mon, 30 Mar 2026 15:33:26 UTC (1,205 KB)
[v2] Mon, 20 Apr 2026 15:05:36 UTC (1,206 KB)
[v3] Tue, 21 Apr 2026 16:58:14 UTC (3,012 KB)
[v4] Fri, 24 Apr 2026 05:06:25 UTC (3,012 KB)
[v5] Thu, 22 Oct 2026 16:51:55 UTC (3,012 KB)
