Loading…
This event has ended. View the official site or create your own event → Check it out
This event has ended. Create your own
View analytic
Tuesday, September 22 • 9:30am - 10:00am
3D Face Tracking and Reconstruction using Modern C++

Log in to save this to your schedule and see who's attending!

In this talk, I will present my work in computer vision, namely landmark detection and 3D face tracking, and the two C++ libraries that were developed in the process. The first part of the talk will give an introduction to detecting facial landmark points and work through a hello-world code example that presents and uses the superviseddescent library, a cross-platform library for cascaded regression that can be used to solve problems like landmark detection or pose estimation. The second part of the talk will present an approach that uses the found landmarks from the first part to use a 3D face model to track a person's face. With the library presented in the process, we hope to make 3D models easier to use and more widespread in the community. Both libraries are designed to be lightweight and simple to use, and try to follow modern C++11/14 programming paradigms. The talk concludes with my views on code sustainability in academia and a wish list of standard library features for computer vision.

Speakers
avatar for Patrik Huber

Patrik Huber

PhD student in computer vision, University of Surrey
I'm a PhD student in computer vision at the Centre for Vision, Speech and Signal Processing of the University of Surrey, UK. My research is about using 3D face models to reconstruct a 3D face from a single 2D image, object tracking and facial landmark detection. My interest in C++ emerged in 2013 when I discovered the power of modern C++ programming, and I have since then open-sourced much of my research software.


Tuesday September 22, 2015 9:30am - 10:00am
Noether (408) Meydenbauer Center