A learning-based view extrapolation method for axial super-resolution

TL;DR

This paper introduces a learning-based approach for axial super-resolution in light fields, enabling more precise refocusing without requiring accurate depth estimation, applicable to various baseline sizes.

Contribution

The proposed method extrapolates novel views from axial volumes of sheared EPIs, improving refocusing precision without the need for depth estimation, suitable for different light field baselines.

Findings

Works well for small baseline light fields like plenoptic cameras

Effective for larger baseline light fields too

Enhances refocusing accuracy with shallower DOF

Abstract

Axial light field resolution refers to the ability to distinguish features at different depths by refocusing. The axial refocusing precision corresponds to the minimum distance in the axial direction between two distinguishable refocusing planes. High refocusing precision can be essential for some light field applications like microscopy. In this paper, we propose a learning-based method to extrapolate novel views from axial volumes of sheared epipolar plane images (EPIs). As extended numerical aperture (NA) in classical imaging, the extrapolated light field gives re-focused images with a shallower depth of field (DOF), leading to more accurate refocusing results. Most importantly, the proposed approach does not need accurate depth estimation. Experimental results with both synthetic and real light fields show that the method not only works well for light fields with small baselines as those captured by plenoptic cameras (especially for the plenoptic 1.0 cameras), but also applies to light fields with larger baselines.