Sparse Photometric 3D Face Reconstruction Guided by Morphable Models

TL;DR

This paper introduces a new 3D face reconstruction method that combines sparse photometric stereo with morphable face models, enabling detailed and accurate facial geometry from minimal images.

Contribution

It develops a robust optimization for lighting calibration and integrates semantic segmentation to enhance detail recovery, surpassing previous methods in quality.

Findings

High-precision lighting calibration without uniform albedo assumption

Reconstruction of fine facial details from few images

Comparable or superior to movie-quality facial detail reconstruction

Abstract

We present a novel 3D face reconstruction technique that leverages sparse photometric stereo (PS) and latest advances on face registration/modeling from a single image. We observe that 3D morphable faces approach provides a reasonable geometry proxy for light position calibration. Specifically, we develop a robust optimization technique that can calibrate per-pixel lighting direction and illumination at a very high precision without assuming uniform surface albedos. Next, we apply semantic segmentation on input images and the geometry proxy to refine hairy vs. bare skin regions using tailored filters. Experiments on synthetic and real data show that by using a very small set of images, our technique is able to reconstruct fine geometric details such as wrinkles, eyebrows, whelks, pores, etc, comparable to and sometimes surpassing movie quality productions.