HDR Environment Map Estimation for Real-Time Augmented Reality

TL;DR

This paper introduces a real-time method for estimating HDR environment maps from narrow FOV LDR images using a neural network, enabling realistic augmented reality reflections and shading.

Contribution

The authors propose EnvMapNet, an efficient neural network architecture with novel loss functions, achieving significant accuracy improvements and real-time performance on mobile devices.

Findings

Reduces directional error of light source estimation by over 50%

Achieves 3.7 times lower FID compared to state-of-the-art methods

Runs in under 9 ms on an iPhone XS for real-time AR rendering

Abstract

We present a method to estimate an HDR environment map from a narrow field-of-view LDR camera image in real-time. This enables perceptually appealing reflections and shading on virtual objects of any material finish, from mirror to diffuse, rendered into a real physical environment using augmented reality. Our method is based on our efficient convolutional neural network architecture, EnvMapNet, trained end-to-end with two novel losses, ProjectionLoss for the generated image, and ClusterLoss for adversarial training. Through qualitative and quantitative comparison to state-of-the-art methods, we demonstrate that our algorithm reduces the directional error of estimated light sources by more than 50%, and achieves 3.7 times lower Frechet Inception Distance (FID). We further showcase a mobile application that is able to run our neural network model in under 9 ms on an iPhone XS, and render in real-time, visually coherent virtual objects in previously unseen real-world environments.