Hi! I’m a collaborator and visiting researcher of the group of Prof. Matthias Rätsch.
Background
I started working in the area of computer vision, and more specifically faces and 3D face models, when I did my B.Sc. thesis under the supervision of Prof. T. Vetter and Prof. Matthias Rätsch in 2008, at University of Basel, Switzerland. After subsequently finishing my M.Sc. with the specialisations Machine Learning and High-performance computing, I pursued a PhD with the topic Real-time 3D Morphable Shape Model Fitting to Monocular In-the-wild Videos at the Centre for Vision, Speech and Signal Processing, of the University of Surrey, United Kingdom. After completing my PhD, I was a Research Fellow (PostDoc) at CVSSP in the FACER2VM project. At the beginning of 2018, I founded 4dface Ltd, a small start-up that has the goal of commercialising 3D morphable face models and 3D face fitting. I am also still a Visiting Academic Researcher at CVSSP, keeping close ties with research and academia. And I love programming in modern C++14/17!
Company website: https://www.4dface.io
Personal research website: https://www.patrikhuber.ch (a bit outdated currently)
Research
As part of 4dface and my research, I’m currently working mainly on two things:
- Building an awesome full-head 3D morphable face model with large expression variety, to enable all kinds of face-related applications in health care, robotics, automotive safety, AR/VR, games and entertainment, marketing research, and many other domains. [More info about the 4D Face Model on 4dface.io/4dfm]
- Improving 3D face tracking and 3D face reconstruction from a single image, as well as from videos and live camera. The goal is to use 3D face information to improve face tracking and face analysis, and be able to create personal 3D face avatars. [More info on 4dface.io/4dface-sdk]
Publications
Recent & highlights:
- Wing Loss for Robust Facial Landmark Localisation With Convolutional Neural Networks
Zhen-Hua Feng, Josef Kittler, Muhammad Awais, Patrik Huber, Xiao-Jun Wu
CVPR 2018, Salt Lake City, USA – [PDF] - 3D Morphable Models as Spatial Transformer Networks
Anil Bas, Patrik Huber, William AP Smith, Muhammad Awais, Josef Kittler
ICCV Workshop on Geometry Meets Deep Learning 2017, Venice, Italy – [PDF] - A Multiresolution 3D Morphable Face Model and Fitting Framework
Patrik Huber, Guosheng Hu, Rafael Tena, Pouria Mortazavian, Willem P Koppen, William Christmas, Matthias Rätsch, Josef Kittler
VISAPP 2016, Rome, Italy – [DOI] [PDF] [Code (GitHub)] - Real-time 3D face super-resolution from monocular in-the-wild videos
Patrik Huber, William Christmas, Adrian Hilton, Josef Kittler, Matthias Rätsch
ACM SIGGRAPH 2016 Posters, Anaheim, USA – [ACM] [Abstract (PDF)] [Poster (PDF)] [YouTube video]
Full list of publications: See my Google Scholar profile.