WaterNeRF: Neural Radiance Fields for Underwater Scenes

Underwater imaging is a critical task performed by marine robots for a wide range of applications including aquaculture, marine infrastructure inspection, and environmental monitoring. However, water column effects, such as attenuation and backscattering, drastically change the color and quality of imagery captured underwater. Due to varying water conditions and range-dependency of these effects, restoring underwater imagery is a challenging problem. This impacts downstream perception tasks including depth estimation and 3D reconstruction. In this paper, we leverage state-of-the-art neural radiance fields (NeRFs) to enable physics-informed novel view synthesis with image restoration and dense depth estimation for underwater scenes. Our proposed method, WaterNeRF, estimates parameters of a physics-based model for underwater image formation and uses these parameters for novel view synthesis. After learning the scene structure and radiance field, we can produce novel views of degraded as well as corrected underwater images. We evaluate the proposed method qualitatively and quantitatively on a real underwater dataset.

I. Introduction

Improving underwater perception will advance autonomous capabilities of underwater vehicles, enabling increased complexity of their missions. This will have impacts across ma-rine science and industry [1], [2]. However, unlike in air, underwater images are often severely degraded due to water column effects on underwater light propagation. In under-water environments, absorption contributes to wavelength-dependent attenuation, resulting in color distortion in the image. Backscattering causes a haze effect across the scene, similar to fog in air [3]. This can pose a problem for marine robotic systems that rely on accurate and consistent color and dense structural information to enable high level perceptual tasks. Ideally, underwater images could be restored to appear as if taken in air. However, water column effects depend on many factors including scene structure, water characteristics, and light properties. Furthermore, there is often a lack of ground truth for color and structure of underwater scenes. These factors make underwater image restoration and dense scene reconstruction an ill-posed problem.