Light Direction and Color Estimation from Single Image with Deep Regression

TL;DR

This paper introduces a deep learning approach to estimate scene light direction and color from a single image, utilizing a synthetic dataset with shadow effects and demonstrating effectiveness on real images.

Contribution

The paper presents a novel deep regression model trained on a synthetic dataset with shadows for light estimation from single images, extending to real scenes.

Findings

Good performance on synthetic images

Effective application to real scenes

New synthetic dataset with shadow effects

Abstract

We present a method to estimate the direction and color of the scene light source from a single image. Our method is based on two main ideas: (a) we use a new synthetic dataset with strong shadow effects with similar constraints to the SID dataset; (b) we define a deep architecture trained on the mentioned dataset to estimate the direction and color of the scene light source. Apart from showing good performance on synthetic images, we additionally propose a preliminary procedure to obtain light positions of the Multi-Illumination dataset, and, in this way, we also prove that our trained model achieves good performance when it is applied to real scenes.