Learning a Deep Convolutional Network for Light-Field Image Super-Resolution

Commercial Light-Field cameras provide spatial and angular information, but its limited resolution becomes an important problem in practical use. In this paper, we present a novel method for Light-Field image super-resolution (SR) via a deep convolutional neural network. Rather than the conventional optimization framework, we adopt a data-driven learning method to simultaneously up-sample the angular resolution as well as the spatial resolution of a Light-Field image. We first augment the spatial resolution of each sub-aperture image to enhance details by a spatial SR network. Then, novel views between the sub-aperture images are generated by an angular super-resolution network. These networks are trained independently but finally fine-tuned via end-to-end training. The proposed method shows the state-of-the-art performance on HCI synthetic dataset, and is further evaluated by challenging real-world applications including refocusing and depth map estimation.