DeepLens: Shallow Depth Of Field From A Single Image

TL;DR

DeepLens introduces a neural network that synthesizes high-resolution shallow depth-of-field images from a single all-in-focus photo, offering flexible focal and aperture control without dual-lens hardware.

Contribution

It presents a novel, fully differentiable neural network model trained on a new RGB-D dataset for high-quality, flexible shallow DoF synthesis from a single image.

Findings

Produces high-resolution shallow DoF images with fewer artifacts.

Achieves results comparable to dual-lens camera portrait mode.

Offers greater flexibility in focal points and aperture settings.

Abstract

We aim to generate high resolution shallow depth-of-field (DoF) images from a single all-in-focus image with controllable focal distance and aperture size. To achieve this, we propose a novel neural network model comprised of a depth prediction module, a lens blur module, and a guided upsampling module. All modules are differentiable and are learned from data. To train our depth prediction module, we collect a dataset of 2462 RGB-D images captured by mobile phones with a dual-lens camera, and use existing segmentation datasets to improve border prediction. We further leverage a synthetic dataset with known depth to supervise the lens blur and guided upsampling modules. The effectiveness of our system and training strategies are verified in the experiments. Our method can generate high-quality shallow DoF images at high resolution, and produces significantly fewer artifacts than the baselines and existing solutions for single image shallow DoF synthesis. Compared with the iPhone portrait mode, which is a state-of-the-art shallow DoF solution based on a dual-lens depth camera, our method generates comparable results, while allowing for greater flexibility to choose focal points and aperture size, and is not limited to one capture setup.