Fast and Full-Resolution Light Field Deblurring using a Deep Neural Network
This work addresses slow processing and limited blur models in light field deblurring, which is essential for parallax-based image processing applications.
The paper tackles the problem of restoring sharp light field images from blurry inputs by generating a complex blurry dataset and proposing a deep learning-based deblurring network, achieving a speedup of about 16,000 times compared to the state-of-the-art method and deblurring full-resolution light fields in under 2 seconds.
Restoring a sharp light field image from its blurry input has become essential due to the increasing popularity of parallax-based image processing. State-of-the-art blind light field deblurring methods suffer from several issues such as slow processing, reduced spatial size, and a limited motion blur model. In this work, we address these challenging problems by generating a complex blurry light field dataset and proposing a learning-based deblurring approach. In particular, we model the full 6-degree of freedom (6-DOF) light field camera motion, which is used to create the blurry dataset using a combination of real light fields captured with a Lytro Illum camera, and synthetic light field renderings of 3D scenes. Furthermore, we propose a light field deblurring network that is built with the capability of large receptive fields. We also introduce a simple strategy of angular sampling to train on the large-scale blurry light field effectively. We evaluate our method through both quantitative and qualitative measurements and demonstrate superior performance compared to the state-of-the-art method with a massive speedup in execution time. Our method is about 16K times faster than Srinivasan et. al. [22] and can deblur a full-resolution light field in less than 2 seconds.