How Many OptiFaces? A New Evaluation Metric For 3D Face Reconstruction

Will Rowan, Patrik Huber, Nick Pears, Andrew Keeling

Keywords: 3D face shape, dataset evaluation, reconstruction metrics, 3D face reconstruction

TL;DR: We present OptiFaces, a novel set of dataset-specific metrics for 3D face reconstruction that provide a more meaningful interpretation of 3D shape reconstruction error.

Abstract: Three dimensional face reconstruction is a challenging problem, so much so that the mean face is highly competitive with recent learning-based approaches for 3D face reconstruction from 2D images. No other universal baselines for this task exist. We propose a novel baseline that selects a subset of face meshes, called OptiFaces, that minimise overall 3D reconstruction error. This is a universal approach to calculate dataset-specific metrics for 3D face reconstruction, offering intuitive new baselines for the interpretation of 3D reconstruction error.

Submission Number: 133