Tree-Mamba: A Tree-Aware Mamba for Underwater Monocular Depth Estimation

2025-07-10

AI Summary

Addressing two core challenges in underwater monocular depth estimation (UMDE)—severe image degradation and unreliable depth supervision—this paper proposes Tree-Mamba, a tree-aware Mamba architecture. First, it introduces a minimum spanning tree (MST) construction method based on feature similarity, coupled with bidirectional (bottom-up and top-down) tree traversal scanning to enhance spatial topological modeling. Second, it establishes BlueDepth, the first high-quality underwater depth benchmark comprising 38,162 image–reliable-depth-map pairs, effectively mitigating object–depth relationship mismatches. Tree-Mamba end-to-end integrates the Mamba backbone with tree-structured feature aggregation. Extensive experiments demonstrate state-of-the-art performance across multiple metrics, with superior qualitative and quantitative results, while maintaining efficient inference speed.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: 3D Computer Vision
• Multiagent Systems: Multiagent Systems under Uncertainty

Application Category

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

Abstract

Underwater Monocular Depth Estimation (UMDE) is a critical task that aims to estimate high-precision depth maps from underwater degraded images caused by light absorption and scattering effects in marine environments. Recently, Mamba-based methods have achieved promising performance across various vision tasks; however, they struggle with the UMDE task because their inflexible state scanning strategies fail to model the structural features of underwater images effectively. Meanwhile, existing UMDE datasets usually contain unreliable depth labels, leading to incorrect object-depth relationships between underwater images and their corresponding depth maps. To overcome these limitations, we develop a novel tree-aware Mamba method, dubbed Tree-Mamba, for estimating accurate monocular depth maps from underwater degraded images. Specifically, we propose a tree-aware scanning strategy that adaptively constructs a minimum spanning tree based on feature similarity. The spatial topological features among the tree nodes are then flexibly aggregated through bottom-up and top-down traversals, enabling stronger multi-scale feature representation capabilities. Moreover, we construct an underwater depth estimation benchmark (called BlueDepth), which consists of 38,162 underwater image pairs with reliable depth labels. This benchmark serves as a foundational dataset for training existing deep learning-based UMDE methods to learn accurate object-depth relationships. Extensive experiments demonstrate the superiority of the proposed Tree-Mamba over several leading methods in both qualitative results and quantitative evaluations with competitive computational efficiency. Code and dataset will be available at https://wyjgr.github.io/Tree-Mamba.html.

Problem

• Estimating depth from degraded underwater images
• Improving Mamba's inflexible state scanning for UMDE
• Addressing unreliable depth labels in UMDE datasets

Innovation

• Tree-aware scanning strategy for feature aggregation
• Minimum spanning tree based on feature similarity
• BlueDepth benchmark with reliable depth labels