WaterGAN: Unsupervised Generative Network to Enable Real-time Color Correction of Monocular Underwater Images

TL;DR

WaterGAN is an unsupervised GAN that generates paired underwater and in-air images to facilitate real-time color correction of monocular underwater images, addressing complex light propagation effects without extensive calibration.

Contribution

The paper introduces WaterGAN, a novel unsupervised GAN framework that creates training data for underwater image color correction, enabling effective deep learning-based restoration without prior water parameter knowledge.

Findings

WaterGAN successfully generates realistic underwater images from in-air images.

The end-to-end network implicitly learns depth estimation from monocular underwater images.

The pipeline improves color correction in real underwater environments.

Abstract

This paper reports on WaterGAN, a generative adversarial network (GAN) for generating realistic underwater images from in-air image and depth pairings in an unsupervised pipeline used for color correction of monocular underwater images. Cameras onboard autonomous and remotely operated vehicles can capture high resolution images to map the seafloor, however, underwater image formation is subject to the complex process of light propagation through the water column. The raw images retrieved are characteristically different than images taken in air due to effects such as absorption and scattering, which cause attenuation of light at different rates for different wavelengths. While this physical process is well described theoretically, the model depends on many parameters intrinsic to the water column as well as the objects in the scene. These factors make recovery of these parameters difficult without simplifying assumptions or field calibration, hence, restoration of underwater images is a non-trivial problem. Deep learning has demonstrated great success in modeling complex nonlinear systems but requires a large amount of training data, which is difficult to compile in deep sea environments. Using WaterGAN, we generate a large training dataset of paired imagery, both raw underwater and true color in-air, as well as depth data. This data serves as input to a novel end-to-end network for color correction of monocular underwater images. Due to the depth-dependent water column effects inherent to underwater environments, we show that our end-to-end network implicitly learns a coarse depth estimate of the underwater scene from monocular underwater images. Our proposed pipeline is validated with testing on real data collected from both a pure water tank and from underwater surveys in field testing. Source code is made publicly available with sample datasets and pretrained models.