About the Project
The Max Planck Institute for Intelligent Systems (MPI‑IS) conducts research to advance intelligent systems that perceive, learn, and interact. Through fundamental research across computational, physical, and social dimensions, we shape artificial intelligence and robotics to advance knowledge and benefit society.
We are offering a:
PhD Position (f/m/div) | Computer Vision
The Optics & Sensing Laboratory (OSLab) at the Max Planck Institute for Intelligent Systems in Tübingen (Germany) has an open position for a Doctoral Researcher for the project "WeGaze: Large-Space Gaze Analysis via Multi-camera System for Social Interactions" to develop datasets and algorithms to capture and analyze eye gaze.
About the project
We are supporting the development of highly realistic human and animal avatars for use in research, film, virtual reality, biology and medicine. Using unique 3D & 4D capture facilities, machine learning, computer vision and advanced graphics, we are modeling humans' and animals' shape and behavior. In the Eye Movement Study, we are closely collaborating with Assistant Professor Xucong Zhang (Delft University of Technology) to develop systems and algorithms that allow for eye gaze study through multiple camera systems.
Laboratory website: https://is.mpg.de/de/optics-sensing-laboratory
Roles & Responsibilities
We are looking for a doctoral researcher to work and develop algorithms and multi-camera capturing systems for eye movements such as gaze, mutual gaze and blinks. The doctoral researcher will be co-supervised by Prof. Xucong Zhang, Delft University of Technology and Dr. Senya Polikovsky, MPI‑IS. This project is funded by the DFG SPP Programme Understanding Gaze (UGaze SPP 2481) and the applicant will be part of the School of SPP Programme as well as the International Max Planck Research School for Intelligent Systems (IMPRS‑IS). The doctoral researcher is expected to work on-site in Delft with Prof. Xucong Zhang for a given time period. They are expected to have strong skills in software design and development as well as experience in computer vision and machine learning.
Your tasks will be to:
- Develop and enhance capture systems which include creating the hardware system and study design.
- Data collection using the developed capture systems
- Develop cutting-edge gaze estimation and blink detection algorithms based on multiple systems in collaboration with TU Delft
- Collaborate with researchers in the field of psychology to adapt and improve the systems for specific case studies
Education & Experience
- MSc degree and strong academic background in the field of computer vision and machine learning
- Strong programming skills including but not limited to Python, OpenCV, PyTorch, Unreal Engine and C++
- Good oral and written communication skills in English
- A natural team player and detail-oriented person
- Some academic background or general interest in psychology would be a plus
Our offer
- Working in a multidisciplinary collaborative project at MPI for Intelligent Systems (Tübingen)
- Cutting-edge technologies related to computer vision and multi-sensor system
- An open-minded and motivating working environment with international scientists, engineers and administrative staff
- We provide a subsidy for the Deutschland-Jobticket (ÖPNV)
- Our connection to a day care center and the possibility of mobile working help you to reconcile family and career.
- The position is for 3 years with the possibility to extend if the scientific progress and funding permit. The position is available as of now.
The Max Planck Society is committed to increasing the number of individuals with disabilities in its workforce and therefore encourages applications from such qualified individuals. The Max Planck Society strives for gender equality and diversity. Furthermore, the Max Planck Society seeks to increase the number of women in its workforce in those areas where they are underrepresented and therefore explicitly encourages women to apply.
Application
Candidates should send their:
- 2-page letter answering:
- What makes you a good fit for this position?
- Download the paper of UniGaze https://ut-vision.github.io/UniGaze/ and answer:
- Why the generic large vision foundation model does not perform best for the gaze estimation task as shown in this paper?
- How would you extend the model to other animals such as apes?
- What are possible issues working with human subject video data and suggestions on how to overcome them?
- What are the pros and cons for using synthetic data for training multi-human gaze detection algorithm? What are the best ways to use synthetic data?
- (Note: figures can also be used in addition to texts)
- Curriculum Vitae (with 3 references)
- (Optional) GitHub or other account(s) showing us your programming skills
Applications without these documents including the letter will not be considered.
Please send the requested documents in a single PDF file, via the application portal.
Any questions regarding the position should be forwarded to Dr. Senya Polikovsky at senya.polikovsky@tuebingen.mpg.de or Dr. Keiko Kitagawa at keiko.kitagawa@tuebingen.mpg.de.
The posting is open until filled.
