项目介绍

2026年7月20日

EAVISE is a multidisciplinary KU Leuven research group based at Campus De Nayer in Sint-Katelijne-Waver. We focus on applications of advanced computer vision, audio and artificial intelligence.
The research goal of EAVISE is to apply state-of-the-art AI, computer vision and image processing and audio techniques as a solution for industry-specific problems. To meet stringent execution speed, energy footprint, cost and price requirements, the developed algorithms are often implemented and optimized on embedded systems, such as GPUs, DSPs and FPGAs.
Application areas include industrial automation, media applications, product inspection, traffic monitoring, e-health, agriculture, eye-tracking, microscopic imaging, surveillance and land surveying.

Website van de eenheid

Project

The media industry has long pursued lifelike digital humans for film and games, and today deploys avatars in virtual worlds, XR, digital twins and immersive training. Yet adoption remains limited: avatars are often perceived as unnatural or robotic. Small inconsistencies in appearance, motion or lighting can trigger uncanny valley effects, undermining acceptance and embodiment. In this project, we explore the possibilities with 3D Gaussian Splatting as an alternative to mesh-based pipelines for photorealistic rendering of digital humans. We aim to develop a novel representation for photorealistic digital humans using 3D Gaussian Splatting as a basis, and to create a coherent, integrated framework for full-body animation-ready 3D Gaussian avatars. This framework will be tested and evaluated on real captured data. 

Profile

• You have a Master’s degree in Electrical Engineering or Computer Science, or a related field that aligns with the topic of the PhD research.
• You have good knowledge of and/or a strong interest in machine learning, computer vision and image processing.
• Research experience in computer vision (e.g. through Master thesis work or research internships) is considered a strong asset.  
• You are creative, take initiative, and can work independently. You are also a team player and are looking forward to work on both theoretical modeling and hands-on experiments.
• You have an excellent CV and are willing to write a personal funding application with the assistance of the supervisors.
• You are proficient in English, have good communication skills and a sense of responsibility.
• You are willing to take up some teaching tasks.

Offer

• A high-level and exciting international research environment.
• Conducting scientific research in the field of machine learning and computer vision.
• Supervising master thesis projects in this research domain.
• Providing teaching assistance in courses depending on the candidate’s profile.
• This is a position in the EAVISE research group. The primary work location for this position is KU Leuven Campus De Nayer (Sint-Katelijne-Waver).

Interested?

For more information please contact Prof. dr. Patrick Vandewalle, mail: patrick.vandewalle@kuleuven.be.You can apply for this job no later than August 31, 2026 via the online application tool