Search papers, labs, and topics across Lattice.
This paper introduces the Degradation-aware Conditional Generation Network (DCGNet) to tackle the challenges of Salient Object Detection (SOD) in underwater images, which are hindered by low contrast and color distortion. The architecture incorporates a Dynamic Multi-Granularity module for scale-variant detection, an Underwater Physics-Prior module for restoring RGB features, and an Underwater Spatial Gaussian module to enhance salient regions while suppressing background clutter. Experimental results on multiple underwater datasets show that DCGNet significantly outperforms existing SOD methods, highlighting its effectiveness in complex underwater visual environments.
DCGNet not only overcomes the limitations of traditional SOD methods in underwater settings but also sets a new benchmark in saliency detection performance.
Salient Object Detection in underwater images remains challenging due to low contrast, uneven illumination, and color distortion caused by scattering and absorption effects, which limit the effectiveness of conventional SOD methods in underwater environments. To address these challenges, we propose a Degradation-aware Conditional Generation Network (DCGNet), specifically designed to construct reliable conditional features for underwater saliency generation. First, we design a Dynamic Multi-Granularity module (DMG) grounded in the human visual system to robustly detect salient objects of varying scales with blurred boundaries. Then, we develop an Underwater Physics-Prior module (UPP), which utilizes pseudo-depth guidance to estimate underwater light attenuation and backscatter, thereby restoring degradation-aware RGB features and mitigating color distortion and boundary ambiguity. Based on the physics-guided representation, we introduce an Underwater Spatial Gaussian module (USG), which constructs a spatial Gaussian saliency prior from the strongest guided response to enhance object-centered salient regions and suppress cluttered underwater backgrounds. In addition, a lightweight timestep-adaptive Diffusion Transformer (DiT) bottleneck is inserted into the denoising decoder to refine fused features at different diffusion timesteps. Comprehensive experiments on USOD10K, USOD, CSOD10K, MAS3K, and RMAS demonstrate that DCGNet significantly outperforms existing state-of-the-art methods, verifying its potential for complex underwater visual applications.
