Visual enhancement and 3D representation for underwater scenes: a review
Guoxi Huang, Haoran Wang, Brett Seymour, Evan Kovacs, John Ellerbrock, Dave Blackham, Nantheera Anantrasirichai
TL;DR
The paper addresses the core challenge of extracting reliable geometry and visually faithful representations from underwater imagery by uniting underwater light transport physics with both traditional and modern learning-based methods. It surveys UVE and underwater 3D reconstruction, contrasting physics-based models (e.g., $I(x)=I_d(x)+I_f(x)+I_b(x)$ and variants like the Jaffe–McGlamery model) with data-driven approaches (CNNs, transformers, diffusion models, NeRF, and 3D Gaussian Splatting), and discusses integrated, end-to-end pipelines. The authors provide a taxonomy of methods, benchmark datasets, and open challenges, including domain adaptation, dynamic scenes, refraction modeling, and real-time rendering, to guide future research and practical deployment. The review highlights that combining physics-informed priors with deep learning yields greater robustness across water types and depths, and points to autonomous large-scale underwater surveys as a key application horizon. Overall, this work serves as a foundational reference for researchers and practitioners aiming to advance underwater exploration through improved visual enhancement and 3D reconstruction techniques.
Abstract
Underwater visual enhancement (UVE) and underwater 3D reconstruction pose significant challenges in computer vision and AI-based tasks due to complex imaging conditions in aquatic environments. Despite the development of numerous enhancement algorithms, a comprehensive and systematic review covering both UVE and underwater 3D reconstruction remains absent. To advance research in these areas, we present an in-depth review from multiple perspectives. First, we introduce the fundamental physical models, highlighting the peculiarities that challenge conventional techniques. We survey advanced methods for visual enhancement and 3D reconstruction specifically designed for underwater scenarios. The paper assesses various approaches from non-learning methods to advanced data-driven techniques, including Neural Radiance Fields and 3D Gaussian Splatting, discussing their effectiveness in handling underwater distortions. Finally, we conduct both quantitative and qualitative evaluations of state-of-the-art UVE and underwater 3D reconstruction algorithms across multiple benchmark datasets. Finally, we highlight key research directions for future advancements in underwater vision.