Sergey Tulyakov

Hi, I am a PhD Candidate under supervision of Nicu Sebe at the Multimedia and Human Understanding Group MHUG at the University of Trento, Italy.

My research topic includes Computer Vision, Machine Learning and 3D Data Processing. Currently, I work with faces: detection, tracking, analysis and reconstruction.

Before joining MHUG group, I had been working as a software engineer for six years.

Contact: sergey.tulyakov@unitn.it

Download CV

My current research is devoted to Unconstrained 2D and 3D Face Analysis, where the term unconstrained means face analysis in realistic situations: outside controlled laboratory setting. I have been working on:

Face detection, tracking and head pose estimation
2D and 3D face alignment
Facial expression recognition
Unconstrained heart rate recognition from videos

Nov 2012 - present:

PhD Canditate at The Univesity of Trento, Italy

Sept 2009 - July 2010:

MSc at the Belarusian State University of Intormatics and Radioelectronics

Sept 2004 - July 2009:

B.Eng at the Belarusian State University of Intormatics and Radioelectronics

Before joining MHUG I worked as a software engineer and project lead for six years:

July 2010 - Sept 2012:

At HiQo-Solutions, Inc I worked as a team and project lead, carrying out architecture design, programming and scientific programming

June 2006 - June 2010:

At Todes, Ltd I worked as a software engineer.

I like creating things that work. It is even more interesting when you can do things with faces. I have put many things I've been working on into the following projects:

FaceCept

FaceCept technology is a set of components that allows real-time analysis of people’s faces. It includes gender, age, facial expression, new/returning and attention time recognition. The technology is cross-platform: it works even in a browser. The project took the first place in ITJUMP 2012.

Here is a demo video:

FaceCept3D

FaceCept3D is a flexible open-source technology for 3D face analysis and recognition, available on GitHub. It allows for head pose estimation and facial expression recognition from extreme head poses. Key advantages of the technology include: flexible architecture that decouples scientific algorithms from technical components and real-time processing pipeline.