Deep Single-Image Portrait Relighting

TLDR

Conventional physically‑based portrait relighting requires solving an inverse rendering problem, but inaccurate estimation of face geometry, reflectance, or lighting often produces strong artifacts that degrade the relit result. This study creates a large‑scale, high‑quality, in‑the‑wild portrait relighting dataset (DPR) by applying a physically‑based relighting method. A deep CNN is trained on DPR to generate relit portraits from a source image and target lighting, with artifact‑removing regularization and a GAN loss, and its performance is evaluated qualitatively and quantitatively on DPR, Flickr, and Multi‑PIE. The trained network relights portraits up to 1024 × 1024 resolution and achieves state‑of‑the‑art results in both qualitative and quantitative tests. Dataset and code are available at https://zhhoper.github.io/dpr.html.

Abstract

Conventional physically-based methods for relighting portrait images need to solve an inverse rendering problem, estimating face geometry, reflectance and lighting. However, the inaccurate estimation of face components can cause strong artifacts in relighting, leading to unsatisfactory results. In this work, we apply a physically-based portrait relighting method to generate a large scale, high quality, "in the wild" portrait relighting dataset (DPR). A deep Convolutional Neural Network (CNN) is then trained using this dataset to generate a relit portrait image by using a source image and a target lighting as input. The training procedure regularizes the generated results, removing the artifacts caused by physically-based relighting methods. A GAN loss is further applied to improve the quality of the relit portrait image. Our trained network can relight portrait images with resolutions as high as 1024 × 1024. We evaluate the proposed method on the proposed DPR datset, Flickr portrait dataset and Multi-PIE dataset both qualitatively and quantitatively. Our experiments demonstrate that the proposed method achieves state-of-the-art results. Please refer to https://zhhoper.github.io/dpr.html for dataset and code.