Personalizing Engagement Detection in Online Learning with Generative AI

Master Thesis: EngageCam

10.11.2025, Abschlussarbeiten, Bachelor- und Masterarbeiten

Are you passionate about advancing AI in online learning? Join the EngageCam project to explore cutting-edge techniques for personalized engagement detection.

Are you passionate about advancing AI in online learning? Join the EngageCam project to explore cutting-edge techniques for personalized engagement detection.

In this project, you'll:
1. Artificially augment datasets by generating synthetic images/videos using Generative AI models, such as StarGAN-v2 (https://github.com/clovaai/stargan-v2), FOMM (https://aliaksandrsiarohin.github.io/first-order-model-website/, https://github.com/alievk/avatarify-python), diffusion-based methods (https://diffusion-motion-transfer.github.io/, https://github.com/showlab/Awesome-Video-Diffusion), or public tools like Viggle (https://viggle.ai/home ). Evaluate model performance through advanced methods, such as similarity measures and engagement/emotion detection using trained ML models.
2. Dive into personalization by:
a. Leveraging federated learning to improve engagement detection accuracy for unseen users, starting with just one image per unseen user.
b. Identifying and utilizing the 3 most similar participants from the dataset (e.g., via k-nearest neighbors, triplet loss) to fine-tune the model for tailored engagement analysis.

Outcome:
Develop a robust, personalized engagement detection system that adapts to unique user needs in online learning environments.

Start Date:
immediately

Datasets:
• EngageNet (https://dl.acm.org/doi/abs/10.1145/3577190.3614164 )
• DAiSEE (https://arxiv.org/abs/1609.01885)
• etc.

Skills Required:
• Advanced knowledge of Machine Learning: Experience with training and evaluating ML models.
• Proficiency in Python: Expertise in using ML libraries such as PyTorch, or scikit-learn.
• Good to have: Experience with Generative AI: Knowledge of techniques for synthetic data generation, such as GANs or diffusion models.

How to apply:
Please write an e-mail to mengdi.wang@tum.de and anna.bodonhelyi@tum.de with a short introduction, CV, and transcripts.

Kontakt: mengdi.wang@tum.de; anna.bodonhelyi@tum.de