Reflection Separation and Deblurring of Plenoptic Images

TL;DR

This paper presents a novel two-stage method for reflection removal and deblurring in plenoptic images, leveraging CNN-based depth estimation and explicit PSF modeling to recover high-resolution textures and sharp scenes.

Contribution

It introduces a combined approach for reflection separation and deblurring in plenoptic images, integrating deep learning and explicit camera modeling for improved results.

Findings

Effective reflection removal demonstrated on real and synthetic images

Successful deblurring of layered scenes with different motion blurs

High-resolution texture recovery achieved through the proposed framework

Abstract

In this paper, we address the problem of reflection removal and deblurring from a single image captured by a plenoptic camera. We develop a two-stage approach to recover the scene depth and high resolution textures of the reflected and transmitted layers. For depth estimation in the presence of reflections, we train a classifier through convolutional neural networks. For recovering high resolution textures, we assume that the scene is composed of planar regions and perform the reconstruction of each layer by using an explicit form of the plenoptic camera point spread function. The proposed framework also recovers the sharp scene texture with different motion blurs applied to each layer. We demonstrate our method on challenging real and synthetic images.