facebook tracking

Master Thesis Project in Computer Vision and Machine Learning

Scroll to content

Tobii is looking for students that wish to explore computer vision and machine learning within eye tracking technology. You will validate state of the art research, come up with new and exciting hypotheses and test them. Above all, your work will impact the way we will experience computer interaction, making virtual reality real and giving people with disability a voice.

You will be part of the Algorithm team that works with the core eye tracking technology. The Algorithm team consists of people with backgrounds within mathematics, computer science, computer vision and machine learning. We develop and research cutting-edge algorithms in computer vision and machine learning for eye and gaze tracking. If you have the competence and passion to explore future technologies of eye and gaze tracking then this project is for you.

The thesis project topic can be chosen from one of the following:

CNN for Realtime Embedded Eye Feature Detection 

Object detection has long been a key challenge in computer vision, involving both object classification and localization. Convolutional neural networks (CNNs) have proven promising in recent years for tackling object detection. These networks however are not suited for deployment on embedded devices with stringent computational and memory requirements. Considerable recent attention has been given to this problem by attempts to reduce the size of existing networks. This project involves using cutting edge techniques to obtain the smallest possible CNN for eye feature detection. 

Learning Colors 

Colorization of images is a popular topic in deep learning, used for example to restore videos and photos in black and white to RGB. This project involves converting black and white images to color using existing sparse color information.

Modeling Feature Detector Errors 

Feature detectors, particularly those that are scaled for real-time use will occasionally produce erroneous output. We would like to model such errors in order to simulate the behaviour of our systems. This thesis topic can be approached from a number of different directions, for example, statistical modelling, generative adversarial networks (GANs), or any other suitable approach.  

We believe you have:

  • Solid background in one or more of the following fields: computer vision, machine learning, mathematics.
  • Good programming skills in C/C++ and/or Python.

Your application should include:

  • CV
  • Cover letter
  • Transcripts

We review applications and interview candidates continuously. Please send your application as soon as possible through our website.

Or, know someone who would be a perfect fit? Let them know!

Stockholm

Karlsrovägen 2D
182 53 Danderyd Directions +46 8 663 69 90

The Tobii way

Tobii is an international company that nurtures a welcoming, friendly atmosphere and non-hierarchical mindset. In Europe, North America and Asia, we cultivate a culture where we all can be ourselves and anyone can talk to anyone, share thoughts and discuss new ideas. This creates an atmosphere where we are more work buddies than colleagues. And the same way that we care for each other on a personal level, the company cares for us as employees; offering work-life balance and great benefits. All this combined spurs career opportunities, innovative ideas and – equally important – fun at work.

Already working at Tobii?

Let’s recruit together and find your next colleague.

  • Ali Farokhian
  • Ali Pandidan
  • Alli Ohms
  • Amanda Bentley
  • Ammar Elyas
  • Anders Frank
  • Anders Wirkestrand
  • Andrei Neculau
  • Andrii Shevchenko
  • Anja Schulze
Teamtailor

Applicant tracking system by Teamtailor