Wu Gaochang, Liu Yebin, Dai Qionghai, Chai Tianyou
IEEE Trans Image Process. 2019 Jul;28(7):3261-3273. doi: 10.1109/TIP.2019.2895463. Epub 2019 Jan 29.
Research in light field reconstruction focuses on synthesizing novel views with the assistance of depth information. In this paper, we present a learning-based light field reconstruction approach by fusing a set of sheared epipolar plane images (EPIs). We start by showing that a patch in a sheared EPI will exhibit a clear structure when the sheared value equals the depth of that patch. By taking advantage of this pattern, a convolutional neural network (CNN) is then trained to evaluate the sheared EPIs, and output a reference score for fusing the sheared EPIs. The proposed CNN is elaborately designed to learn the similarity degree between the input sheared EPI and the ground truth EPI. Therefore, no depth information is required for network training and reasoning. We demonstrate the high performance of the proposed method through evaluations on synthetic scenes, real-world scenes, and challenging microscope light fields. We also show a further application of our proposed network for depth inference.
光场重建的研究主要集中在利用深度信息合成新视图。在本文中，我们提出了一种基于学习的光场重建方法，该方法通过融合一组剪切极平面图像（EPI）来实现。我们首先表明，当剪切值等于该补丁的深度时，剪切EPI中的一个补丁将呈现出清晰的结构。利用这种模式，然后训练一个卷积神经网络（CNN）来评估剪切EPI，并输出一个用于融合剪切EPI的参考分数。所提出的CNN经过精心设计，以学习输入剪切EPI与真实EPI之间的相似度。因此，网络训练和推理不需要深度信息。我们通过对合成场景、真实世界场景和具有挑战性的显微镜光场进行评估，证明了所提方法的高性能。我们还展示了所提网络在深度推理方面的进一步应用。
