OpenWaters: Photorealistic Simulations For Underwater Computer Vision
OpenWaters: Photorealistic Simulations For Underwater Computer Vision
复制标题
DOI：
10.1145/3491315.3491336
复制
发表时间：
2021-11
期刊：
影响因子：
--
通讯作者：
Mehdi Mousavi;Shardul Vaidya;Razat Sutradhar;A. Ashok
中科院分区：
文献类型：
--
作者：
Mehdi Mousavi;Shardul Vaidya;Razat Sutradhar;A. Ashok
本文介绍了OpenWaters，这是一个实时开源的水下仿真工具包，用于生成逼真的水下场景。OpenWaters通过模拟不同的现实世界条件，支持创建大量的水下图像。它允许对模拟实例中的每个变量进行精细控制，包括几何体、渲染参数，如光线跟踪水焦散、散射和地面真实值标签。以水下深度(摄像机与物体之间的距离)估计为例，展示并验证了OpenWaters对水下场景建模的能力，该能力用于训练深度神经网络进行深度估计。我们的实验评估表明，使用高精度的合成水下图像进行深度估计是可行的，并且可以将合成图像中的特征迁移学习到真实世界的图像中。
In this paper, we present OpenWaters, a real-time open-source underwater simulation kit for generating photorealistic underwater scenes. OpenWaters supports creation of massive amount of underwater images by emulating diverse real-world conditions. It allows for fine controls over every variable in a simulation instance, including geometry, rendering parameters like ray-traced water caustics, scattering, and ground-truth labels. Using underwater depth (distance between camera and object) estimation as the use-case, we showcase and validate the capabilities of OpenWaters to model underwater scenes that are used to train a deep neural network for depth estimation. Our experimental evaluation demonstrates depth estimation using synthetic underwater images with high accuracy, and feasibility of transfer-learning of features from synthetic to real-world images.