Wang Xucheng, Tao Chenning, Wu Rengmao, Tao Xiao, Sun Peng, Li Yong, Zheng Zhenrong
J Opt Soc Am A Opt Image Sci Vis. 2020 Jul 1;37(7):1236-1243. doi: 10.1364/JOSAA.388555.
In this paper, we propose a convolutional neural network based on epipolar geometry and image segmentation for light-field depth estimation. Epipolar geometry is utilized to estimate the initial disparity map. Multi-orientation epipolar images are selected as input data, and the convolutional blocks are adopted based on the disparity of different-direction epipolar images. Image segmentation is used to obtain the edge information of the central sub-aperture image. By concatenating the output of the two parts, an accurate depth map could be generated with fast speed. Our method achieves a high rank on most quality assessment metrics in the HCI 4D Light Field Benchmark and also shows effectiveness in estimating accurate depth on real-world light-field images.
在本文中，我们提出了一种基于对极几何和图像分割的卷积神经网络，用于光场深度估计。利用对极几何来估计初始视差图。选择多方向对极图像作为输入数据，并基于不同方向对极图像的视差采用卷积块。使用图像分割来获取中央子孔径图像的边缘信息。通过将两部分的输出连接起来，可以快速生成准确的深度图。我们的方法在HCI 4D光场基准测试的大多数质量评估指标上都取得了很高的排名，并且在估计真实世界光场图像的准确深度方面也显示出有效性。
