Multi-view Consensus CNN for 3D Facial Landmark Placement

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedings – Annual report year: 2019Researchpeer-review

View graph of relations

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.
Original languageEnglish
Title of host publicationProceedings of 14th Asian Conference on Computer Vision
PublisherSpringer
Publication date2019
Pages706-719
ISBN (Print)9783030208875
DOIs
Publication statusPublished - 2019
Event14th Asian Conference on Computer Vision - Perth Convention and Exhibition Centre , Perth, Australia
Duration: 2 Dec 20186 Dec 2018

Conference

Conference14th Asian Conference on Computer Vision
LocationPerth Convention and Exhibition Centre
CountryAustralia
CityPerth
Period02/12/201806/12/2018
SeriesLecture Notes in Computer Science
Volume11361
ISSN0302-9743
CitationsWeb of Science® Times Cited: No match on DOI

    Research areas

  • 3D facial landmarks, Multi-view CNN, Geometric deep learning

ID: 180161480