Face identification across different poses and illuminations with a 3D morphable model

TLDR

The authors propose a novel approach for recognizing faces in images taken from different directions and under varying illumination. The method employs a 3D morphable face model that encodes shape and texture as parameters and an algorithm that recovers these parameters from a single image, separating them from pose and illumination for identification. The system outputs identity with a confidence measure and demonstrates performance on more than 4000 images from the CMU‑PIE database.

Abstract

We present a novel approach for recognizing faces in images taken from different directions and under different illumination. The method is based on a 3D morphable face model that encodes shape and texture in terms of model parameters, and an algorithm that recovers these parameters from a single image of a face. For face identification, we use the shape and texture parameters of the model that are separated from imaging parameters, such as pose and illumination. In addition to the identity, the system provides a measure of confidence. We report experimental results for more than 4000 images from the publicly available CMU-PIE database.