PostDoc Project in Real-Time Visual Computing and AI

KU - SCIENCE - DATALOGISK INSTITUT - UP1; København; On-site; DKK 355,000 - 408,000; Full time

Job summary

The Postdoc position at KU - Science - Datalogisk Institut (UP1) invites applications for REVCAI, focused on real-time visual computing and AI. The role emphasizes efficient algorithms, high-performance computing, and research contributions across graphics, vision, and machine learning.

The successful candidate will develop novel methods, supervise PhD students, and collaborate with international partners, with a starting date around March 1, 2027.

Qualifications

• PhD in relevant field with strong research record.
• Experience with real-time visual computing and AI is advantageous.
• Strong programming and algorithmic problem-solving skills.

Responsibilities

• Carry through an independent research project under supervision.
• Supervise and support prospective PhD students.
• Participate in international collaboration and dissemination of results.
• Publish in leading conferences and journals.

Skills

Python

C++

Graphics & Vision

Machine Learning

High-Performance Computing

Algorithm Design

Education

PhD in Computer Science / Engineering / Robotics / Applied Mathematics

Tools

Python

C++

GPU Programming

Job description

PostDoc Project in Real-Time Visual Computing and AI

The Department of Computer Science invites applicants for a Postdoc position in Real-Time Visual Computing and AI (REVCAI).

REVCAI investigates how advances in visual computing and artificial intelligence can be translated into systems that are not only accurate, but also fast, efficient, scalable, responsive, and practical to deploy. Modern AI systems have made remarkable progress in understanding and generating visual information, but their computational requirements, latency, limited adaptability, and dependence on large-scale computing infrastructure remain major obstacles to their use in interactive and real-world settings.

The project will develop new algorithms and computational methods at the intersection of computer graphics, computer vision, machine learning, geometry, and high-performance computing. Particular emphasis will be placed on representations and algorithms that make sophisticated visual intelligence possible under practical computational constraints.

The expected start date is March 1, 2027, or as soon as possible thereafter.

The Postdoc will investigate fundamental problems in real-time and resource-efficient visual computing and AI. The precise research direction will be developed together with the successful candidate and may build on their particular strengths and interests.

Potential research directions include real-time 3D reconstruction and representation, neural and point-based representations, efficient rendering and visualisation, high-performance computer vision, efficient machine learning, dynamic and multimodal scene understanding, geometric and physical reasoning, and visual perception for interactive or embodied systems.

A central theme of the project is that computational efficiency is not merely an implementation detail. Latency, memory consumption, scalability, and computational cost can fundamentally determine whether an algorithm can be used in an interactive system, on resource-constrained hardware, or as part of a larger perception-and-action loop. The project will therefore investigate algorithmic and representational choices that improve computational performance while retaining - or improving - accuracy, robustness, and generalisation.

Depending on the research direction, the work may range from low-level GPU and systems optimisation to the design of new representations, learning algorithms, or complete real-time visual computing systems. Applications may include robotics and autonomous systems, interactive 3D applications, augmented and virtual reality, scientific and biomedical imaging, and other domains in which visual information must be processed efficiently and reliably.

The successful candidate will be expected to develop novel methods, implement and experimentally evaluate them, compare them against relevant state of the art approaches, as well as supervise PhD students or undergraduate students who carry out the above tasks. We are particularly interested in research that combines strong algorithmic ideas with convincing practical demonstrations and careful empirical evaluation.

Who are we looking for?

We are looking for candidates who have completed their PhD within the fields of Computer Science, Computer Engineering, Robotics, Applied Mathematics, or related areas.

Applicants should have strong technical and analytical skills and an interest in developing high-performance algorithms for visual computing and artificial intelligence.

Experience in one or more of the following areas is advantageous:

• computer graphics or computer vision,
• machine learning,
• 3D reconstruction or representation,
• GPU programming or high-performance computing,
• robotics or embodied AI,
• geometry processing,
• scientific computing,
• optimisation or applied mathematics.

What do we offer?

The project is supervised by Prof. Francois Lauze and Assistant Professor Bernhard Kerbl at the Department of Computer Science, University of Copenhagen.

The postdoc will join the IMAGE section at the University of Copenhagen, an active research environment working across computer graphics, computer vision, image analysis, geometry, computational modelling, and artificial intelligence.

The project will provide substantial freedom to develop an independent research profile within the broader REVCAI agenda. The postdoc will have opportunities to collaborate with researchers at the University of Copenhagen and with national and international research partners, and to contribute both to scientific publications and to open-source research software.

IMAGE offers a friendly, collaborative, and informal research environment with close interaction between faculty, postdoctoral researchers, and PhD students. We value curiosity, technical depth, openness, and teamwork, and aim to provide a supportive environment.

Principal supervisor: Francois Lauze, Department of Computer Science

Contact E-mail: [e-mail]

Contact Telephone: [phone]

The employment as Postdoc is a full-time position for 2 years. Starting date is November 15, 2026, or after agreement.

The position is covered by the Memorandum on Job Structure for Academic Staff.

Salary, pension, and other conditions of employment are set in accordance with the Agreement between the Ministry of Finance and AC (Danish Confederation of Professional Associations) or other relevant organisation. Currently, the monthly salary starts at 31,800 DKK/approx. 4,200 EUR (August 2026 level). Depending on qualifications, a supplement may be negotiated.

Responsibilities:

• Carry through an independent research project under supervision.
• Supervise and support prospective PhD students
• Participate actively in research environments, including a stay at another research institution, preferably abroad.
• Participate in teaching and knowledge-dissemination activities.
• Write scientific papers aimed at leading international conferences and journals.

We are looking for the following qualifications:

• A strong academic background in computer science, computer engineering, robotics, applied mathematics, physics, or a related field.
• Strong programming and algorithmic problem-solving skills.
• Experience with one or more of computer graphics, computer vision, machine learning, high-performance computing, robotics, geometry processing, optimisation, or related quantitative methods.
• Relevant research, thesis, project, or publication experience.
• Programming experience, preferably in Python and/or C++.
• An interest in computational efficiency and in understanding the algorithmic and systems aspects of modern visual AI.
• A curious and analytical mindset and an ability to work independently as well as collaboratively.
• Good written and spoken English skills.

Application deadline

The deadline for applications is 6 November 2027, 23:59 CET.