# Underwater Monocular Metric Depth Estimation: Real-World Benchmarks and Synthetic Fine-Tuning with Vision Foundation Models

arXiv: https://arxiv.org/abs/2507.02148 · HTML: https://mlsanity.com/paper/2507.02148/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 37 of its 78 references that are in the map.
- First posted on arXiv 2 Jul 2025.

## Builds on (most cited first, 30 of 37)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [SGDR: Stochastic Gradient Descent with Warm Restarts](https://mlsanity.com/paper/1608.03983/) (2016)
- [Vision Transformers for Dense Prediction](https://mlsanity.com/paper/2103.13413/) (2021)
- [Depth Anything V2](https://mlsanity.com/paper/2406.09414/) (2024)
- [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://mlsanity.com/paper/2401.10891/) (2024)
- [Depth Map Prediction from a Single Image using a Multi-Scale Deep Network](https://mlsanity.com/paper/1406.2283/) (2014)
- [Diffusion Models in Vision: A Survey](https://mlsanity.com/paper/2209.04747/) (2022)
- [Fine-Tuning can Distort Pretrained Features and Underperform Out-of-Distribution](https://mlsanity.com/paper/2202.10054/) (2022)
- [ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders](https://mlsanity.com/paper/2301.00808/) (2023)
- [ZoeDepth: Zero-shot Transfer by Combining Relative and Metric Depth](https://mlsanity.com/paper/2302.12288/) (2023)
- [Hypersim: A Photorealistic Synthetic Dataset for Holistic Indoor Scene Understanding](https://mlsanity.com/paper/2011.02523/) (2020)
- [How to train your ViT? Data, Augmentation, and Regularization in Vision Transformers](https://mlsanity.com/paper/2106.10270/) (2021)
- [Pyramid Stereo Matching Network](https://mlsanity.com/paper/1803.08669/) (2018)
- [Depth Pro: Sharp Monocular Metric Depth in Less Than a Second](https://mlsanity.com/paper/2410.02073/) (2024)
- [Metric3Dv2: A Versatile Monocular Geometric Foundation Model for Zero-shot Metric Depth and Surface Normal Estimation](https://mlsanity.com/paper/2404.15506/) (2024)
- [Unified Deep Supervised Domain Adaptation and Generalization](https://mlsanity.com/paper/1709.10190/) (2017)
- [When Vision Transformers Outperform ResNets without Pre-training or Strong Data Augmentations](https://mlsanity.com/paper/2106.01548/) (2021)
- [An Underwater Image Enhancement Benchmark Dataset and Beyond](https://mlsanity.com/paper/1901.05495/) (2019)
- [UniDepthV2: Universal Monocular Metric Depth Estimation Made Simpler](https://mlsanity.com/paper/2502.20110/) (2025)
- [Surrogate Gap Minimization Improves Sharpness-Aware Training](https://mlsanity.com/paper/2203.08065/) (2022)
- [A Survey of Structure from Motion](https://mlsanity.com/paper/1701.08493/) (2017)
- [Underwater Single Image Color Restoration Using Haze-Lines and a New Quantitative Dataset](https://mlsanity.com/paper/1811.01343/) (2018)
- [Indoor Semantic Segmentation using depth information](https://mlsanity.com/paper/1301.3572/) (2013)
- [Domain Adaptation for Underwater Image Enhancement](https://mlsanity.com/paper/2108.09650/) (2021)
- [Structure-Preserving Multi-Domain Stain Color Augmentation using Style-Transfer with Disentangled Representations](https://mlsanity.com/paper/2107.12357/) (2021)
- [Atlantis: Enabling Underwater Depth Estimation with Stable Diffusion](https://mlsanity.com/paper/2312.12471/) (2023)
- [UDepth: Fast Monocular Depth Estimation for Visually-guided Underwater Robots](https://mlsanity.com/paper/2209.12358/) (2022)
- [SyreaNet: A Physically Guided Underwater Image Enhancement Framework Integrating Synthetic and Real Images](https://mlsanity.com/paper/2302.08269/) (2023)

## Abstract (arXiv)

Monocular depth estimation has recently progressed beyond ordinal depth to provide metric depth predictions. However, its reliability in underwater environments remains limited due to light attenuation and scattering, color distortion, turbidity, and the lack of high-quality metric ground truth data. In this paper, we present a comprehensive benchmark of zero-shot and fine-tuned monocular metric depth estimation models on real-world underwater datasets with metric depth annotations, including FLSea and SQUID. We evaluated a diverse set of state-of-the-art Vision Foundation Models across a range of underwater conditions and depth ranges. Our results show that large-scale models trained on terrestrial data (real or synthetic) are effective in in-air settings, but perform poorly underwater due to significant domain shifts. To address this, we fine-tune Depth Anything V2 with a ViT-S backbone encoder on a synthetic underwater variant of the Hypersim dataset, which we simulated using a physically based underwater image formation model. Our fine-tuned model consistently improves performance across all benchmarks and outperforms baselines trained only on the clean in-air Hypersim dataset. This study presents a detailed evaluation and visualization of monocular metric depth estimation in underwater scenes, emphasizing the importance of domain adaptation and scale-aware supervision for achieving robust and generalizable metric depth predictions using foundation models in challenging environments.