DualPhys-GS: Dual Physically-Guided 3D Gaussian Splatting for Underwater Scene Reconstruction

2025-08-13

AI Summary

Underwater 3D reconstruction suffers from severe color distortion, geometric artifacts, and structural collapse due to wavelength-selective absorption and scattering by suspended particles. Method: This paper proposes a dual physics-guided 3D Gaussian splatting framework. It introduces two parallel optimization pathways: RGB-guided wavelength attenuation modeling and multi-scale depth-aware scattering modeling, integrated with adaptive water-type classification and dynamic parameter adjustment. The architecture incorporates a feature pyramid network, attention mechanisms, and three novel losses—edge-aware scattering loss, multi-scale feature loss, and physics-consistency constraint loss. Results: Extensive experiments demonstrate that our method significantly outperforms state-of-the-art approaches in turbid waters and at long distances, achieving substantial improvements in reconstruction accuracy and visual fidelity. It effectively mitigates chromatic distortion and geometric collapse, and—uniquely—achieves synergistic enhancement of physical interpretability and neural rendering performance.

Technology Category

• Computer Vision: 3D Computer Vision
• Search and Optimization: Mixed Discrete/Continuous Search
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Sustainability and carbon-aware systems for Web, mobile, and WoT

Abstract

In 3D reconstruction of underwater scenes, traditional methods based on atmospheric optical models cannot effectively deal with the selective attenuation of light wavelengths and the effect of suspended particle scattering, which are unique to the water medium, and lead to color distortion, geometric artifacts, and collapsing phenomena at long distances. We propose the DualPhys-GS framework to achieve high-quality underwater reconstruction through a dual-path optimization mechanism. Our approach further develops a dual feature-guided attenuation-scattering modeling mechanism, the RGB-guided attenuation optimization model combines RGB features and depth information and can handle edge and structural details. In contrast, the multi-scale depth-aware scattering model captures scattering effects at different scales using a feature pyramid network and an attention mechanism. Meanwhile, we design several special loss functions. The attenuation scattering consistency loss ensures physical consistency. The water body type adaptive loss dynamically adjusts the weighting coefficients. The edge-aware scattering loss is used to maintain the sharpness of structural edges. The multi-scale feature loss helps to capture global and local structural information. In addition, we design a scene adaptive mechanism that can automatically identify the water-body-type characteristics (e.g., clear coral reef waters or turbid coastal waters) and dynamically adjust the scattering and attenuation parameters and optimization strategies. Experimental results show that our method outperforms existing methods in several metrics, especially in suspended matter-dense regions and long-distance scenes, and the reconstruction quality is significantly improved.

Problem

• Addresses color distortion in underwater 3D reconstruction
• Mitigates geometric artifacts from light scattering
• Improves long-distance scene reconstruction accuracy

Innovation

• Dual-path optimization for underwater reconstruction
• Dual feature-guided attenuation-scattering modeling
• Scene adaptive mechanism for water-body-type