Unsupervised Learning Based Focal Stack Camera Depth Estimation

TL;DR

This paper introduces an unsupervised deep learning approach for estimating depth from focal stack images, outperforming single-image methods on the NYU-v2 dataset.

Contribution

It presents a novel unsupervised learning framework specifically designed for depth estimation from focal stacks, improving accuracy over existing single-image techniques.

Findings

Significantly better depth accuracy on NYU-v2 dataset

Effective unsupervised learning approach for focal stack depth estimation

Outperforms traditional single-image based methods

Abstract

We propose an unsupervised deep learning based method to estimate depth from focal stack camera images. On the NYU-v2 dataset, our method achieves much better depth estimation accuracy compared to single-image based methods.