Multi-view consensus CNN for 3D facial landmark placement

TL;DR

This paper introduces a multi-view CNN approach that leverages geometric deep learning and human pose tracking techniques to improve the accuracy of 3D facial landmark placement, outperforming existing methods.

Contribution

It combines multi-view CNNs with pose tracking and consensus methods for enhanced 3D facial landmark detection, bridging 2D estimates with 3D space.

Findings

Outperforms current methods significantly on standard datasets

Effective transfer from 3D scans to MRI landmark placement

Demonstrates the utility of geometric deep learning in 3D face analysis

Abstract

The rapid increase in the availability of accurate 3D scanning devices has moved facial recognition and analysis into the 3D domain. 3D facial landmarks are often used as a simple measure of anatomy and it is crucial to have accurate algorithms for automatic landmark placement. The current state-of-the-art approaches have yet to gain from the dramatic increase in performance reported in human pose tracking and 2D facial landmark placement due to the use of deep convolutional neural networks (CNN). Development of deep learning approaches for 3D meshes has given rise to the new subfield called geometric deep learning, where one topic is the adaptation of meshes for the use of deep CNNs. In this work, we demonstrate how methods derived from geometric deep learning, namely multi-view CNNs, can be combined with recent advances in human pose tracking. The method finds 2D landmark estimates and propagates this information to 3D space, where a consensus method determines the accurate 3D face landmark position. We utilise the method on a standard 3D face dataset and show that it outperforms current methods by a large margin. Further, we demonstrate how models trained on 3D range scans can be used to accurately place anatomical landmarks in magnetic resonance images.