The e-Media Research Lab is a research section of Dynamical Systems, Signal Processing, and Data Analytics (STADIUS) in the Department of Electrical Engineering (ESAT) at KU Leuven. The e-Media Lab''s research includes topics related to signal processing, data analysis, machine learning, and Human-Computer Interaction (HCI). E-Media made significant contributions to critical applications in the domains of healthcare, Industry 5.0, biomedical sensing, and education. In particular, e-Media is exploring novel tinyML technologies to enhance the intelligence and energy efficiency of resource-constrained devices.
Project
Digital imagery has become fundamental to smartphones, surveillance, healthcare, and smart‑city systems. AI vision tools can recognize faces, detect objects, track people, and generate synthetic images with ease. These capabilities also introduce major privacy and ethical risks, amplified by recent deepfake misuse and strict regulations such as GDPR.
We aim to develop a real‑time, privacy‑preserving camera system that anonymizes individuals directly on the device. Instead of blurring, it replaces each face with a natural‑looking synthetic version that preserves gender, age, expression, and gaze while concealing identity. Raw video never leaves the device, and authorized reversal is possible when legally required. With such a privacy-safe camera, numerous potential applications emerge, such as privacy-preserving visual sensing in healthcare settings or anonymized person detection in surveillance systems.
To achieve this goal, we are looking for an enthusiastic PhD researcher who will focus on bringing face-swapping algorithms to off-the-shelf embedded platforms. The work will involve deploying the face de‑identification pipeline on resource‑constrained hardware, optimizing the underlying AI models based on latency, memory, and power constraints, and building demonstrators that showcase real‑time anonymization in practice. The researcher will develop hardware‑aware neural architectures with hardware-in-the-loop optimization and efficient inference strategies to meet the computational demands of real-time, edge-based processing.
Profile
We are looking for a highly motivated PhD researcher with an interest in optimizing AI models on resource-constrained embedded hardware. Requirements include:
- Master’s degree in Electrical Engineering
- Ranked within the top 10% of their class in MSc and BSc, and have exceptional grades
- Good background in deep learning with familiarity in model training, inference, and evaluation
- Experience with embedded hardware, such as AI accelerators (e.g., NVIDIA Jetson, Axelera, Qualcomm, Rockchip) or ARM-based platforms
- Solid programming skills, particularly in Python and C/C++, and experience with deep learning frameworks such as PyTorch or TensorFlow
- Interest in hardware aware AI, including model compression, quantization, pruning, or efficient neural architectures for low-power devices
- Strong communication skills and be fluent in English
- Excellent interpersonal skills and the ability to work in an international team
Offer
We offer:
- A PhD scholarship for up to four years (subject to positive intermediate evaluations). If the applicant is non-EER, up to 1 year of pre-doc will also be supported
- An exciting research environment, working on the intersection between theory and implementation in a very multi-disciplinary research environment
- A PhD title from a highly ranked university (#1 in Europe in terms of scientific innovation) after approximately 4 years of successful research
- A thorough scientific education, the possibility of becoming a world-class researcher
- A KU Leuven affiliation, one of the largest research universities in Europe
- The possibility to participate in international conferences, workshops and collaborations with top EU research teams
Starting date: from October 1st 2026 (to be discussed)
Interested?
To apply, please submit the following in the online portal:
A CV mentioning BSc and MSc average grades.
A motivation letter (1 page maximum)
Complete transcripts of Bachelor and Master Degrees (if the MSc degree is not finished yet, please send the most recent transcript).
(if available) IELTS, TOEFL or similar test proving your proficiency in English.
For more information please contact Prof. Dr. ir. Jona Beysens, mail: jona.beysens@kuleuven.be.
KU Leuven strives for an inclusive, respectful and socially safe environment. We embrace diversity among individuals and groups as an asset. Open dialogue and differences in perspective are essential for an ambitious research and educational environment. In our commitment to equal opportunity, we recognize the consequences of historical inequalities. We do not accept any form of discrimination based on, but not limited to, gender identity and expression, sexual orientation, age, ethnic or national background, skin colour, religious and philosophical diversity, neurodivergence, employment disability, health, or socioeconomic status. For questions about accessibility or support offered, we are happy to assist you at this email address.
Fields: Signal Processing, Artificial Intelligence, Machine Learning, Electrical Engineering
