High-Resolution Underwater Camouflaged Object Detection: GBU-UCOD Dataset and Topology-Aware and Frequency-Decoupled Networks

2026-02-03

AI Summary

This work addresses the challenging detection of camouflaged deep-sea objects, which exhibit high visual similarity to their background, weak transparency cues, and fragile elongated structures prone to fragmentation. To tackle these issues, we propose DeepTopo-Net, a novel framework integrating topological awareness with frequency-domain decoupling. The approach incorporates a Water Condition Adaptive Perceiver (WCAP) and an Abyssal Topology Refinement Module (ATRM), complemented by Riemannian metric tensor-guided adaptive sampling and a skeleton-prior topology-preserving strategy to effectively maintain structural integrity of slender targets. We also introduce GBU-UCOD, the first 2K-resolution dataset tailored to vertically stratified marine environments. Extensive experiments demonstrate state-of-the-art performance on MAS3K, RMAS, and GBU-UCOD, significantly advancing detection accuracy and morphological fidelity for camouflaged objects in complex underwater scenes.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Underwater Camouflaged Object Detection (UCOD) is a challenging task due to the extreme visual similarity between targets and backgrounds across varying marine depths. Existing methods often struggle with topological fragmentation of slender creatures in the deep sea and the subtle feature extraction of transparent organisms. In this paper, we propose DeepTopo-Net, a novel framework that integrates topology-aware modeling with frequency-decoupled perception. To address physical degradation, we design the Water-Conditioned Adaptive Perceptor (WCAP), which employs Riemannian metric tensors to dynamically deform convolutional sampling fields. Furthermore, the Abyssal-Topology Refinement Module (ATRM) is developed to maintain the structural connectivity of spindly targets through skeletal priors. Specifically, we first introduce GBU-UCOD, the first high-resolution (2K) benchmark tailored for marine vertical zonation, filling the data gap for hadal and abyssal zones. Extensive experiments on MAS3K, RMAS, and our proposed GBU-UCOD datasets demonstrate that DeepTopo-Net achieves state-of-the-art performance, particularly in preserving the morphological integrity of complex underwater patterns. The datasets and codes will be released at https://github.com/Wuwenji18/GBU-UCOD.

Problem

• Underwater Camouflaged Object Detection
• topological fragmentation
• transparent organisms
• visual similarity
• marine depth

Innovation

• topology-aware modeling
• frequency-decoupled perception
• Riemannian metric tensors
• skeletal priors
• high-resolution underwater dataset