Deep Sparse Light Field Refocusing
This work addresses the hardware and performance limitations in light field imaging, such as bulky structures and resolution trade-offs, by enabling compact devices and better compressive systems for applications in photography and computational imaging.
The paper tackles the problem of light field refocusing from sparse angular views, which traditionally requires dense angular sampling, by proposing a novel convolutional neural network that enables fast, memory-efficient reconstruction without retraining for different refocusing ranges and noise levels, achieving major improvements over existing methods.
Light field photography enables to record 4D images, containing angular information alongside spatial information of the scene. One of the important applications of light field imaging is post-capture refocusing. Current methods require for this purpose a dense field of angle views; those can be acquired with a micro-lens system or with a compressive system. Both techniques have major drawbacks to consider, including bulky structures and angular-spatial resolution trade-off. We present a novel implementation of digital refocusing based on sparse angular information using neural networks. This allows recording high spatial resolution in favor of the angular resolution, thus, enabling to design compact and simple devices with improved hardware as well as better performance of compressive systems. We use a novel convolutional neural network whose relatively small structure enables fast reconstruction with low memory consumption. Moreover, it allows handling without re-training various refocusing ranges and noise levels. Results show major improvement compared to existing methods.