Chair for Human-Centered Artificial Intelligence

  1. University
  2. Faculties
  3. Faculty of Applied Computer Science
  4. Institute of Computer Science
  5. Chairs and Professors
  6. Human-Centered Artificial Intelligence
  7. Team
  8. Daksitha Withanage Don

Daksitha Withanage Don M.Sc.

Research Associate
Chair for Human-Centered Artificial Intelligence
Phone: +49 821 – 598 2305
Email:
Room: 2044 (N)
Address: Universitätsstraße 6a, 86159 Augsburg

Research Interests

  • Affective Computing
  • Artificial Emotional Intelligence
  • Socially Interactive Agents
  • Generative AI 
  • Self-supervised Learning

Bachelor/Master Thesis or Project Module

Thesis Guidelines for Prospective Students

If you're interested in writing a Bachelor’s or Master’s thesis with me, please follow these guidelines:

Eligibility

You have a Computer science related background and prior experience in topics related to my research areas is an advantage.

That is great if you have taken one of our courses:

  • Generative AI for HCI
  • Human-Computer Interaction
  • Bachelor/Master seminars in Generative AI

How to Apply

1. Looking for a Topic

  • Explore the Open Topics section below.
  • If no topics match your interests, you can propose your own by emailing:
    • A motivational statement explaining why the topic fits your interests.
    • Your transcript of records (current and previous, if applicable).
    • A timeframe for the thesis (planned start and end dates).
    • Note: Supervision depends on my capacity and topic relevance.

2. Have Your Own Topic or External Proposal

  • If proposing your own or an external company topic:
    • Include how it aligns with my research.
    • For company topics, attach the original description and any specific requirements (e.g., NDAs).
    • If suitable, I will guide you through the next steps.

Next Steps

  • If Accepted:

    • We will formalize your topic and discuss project goals.
    • Ensure you meet university-specific requirements (e.g., registration, defense talks).

  • If Declined:

    • You may revise your topic or explore alternative supervisors.

Evaluation Criteria

Your thesis will be graded on:

  • Literature review.
  • Scientific approach and methodology.
  • Clear structure and comprehensive documentation.
  • Novelty and significance (especially for master students).
  • Quality of implementation or study design.

Feel free to contact me for further clarification or to apply. Looking forward to working on exciting projects together!

Open Topics :
 

GUI Design for a Social Interactive Agent Framework
Bachelor/Master
This thesis focuses on designing and implementing a user-friendly interface for a Python-based framework using tools like Qt Designer. The work will incorporate best practices in UI/UX design and include an introduction to generative models such as text-to-speech, speech-to-text, and APIs like OpenAI or Llama. Prior knowledge of these technologies is beneficial but not required.
 
 
Behavioral Synchrony Using Vision and Audio Foundation Models
Bachelor/Master
Foundational models are large, pre-trained neural networks capable of processing diverse data types like video (DinoV2), audio (Wav2Vec2_BERT), and text (GPT4), making them highly suitable for behavior analysis. Unlike traditional methods that rely on manual feature engineering, foundational models extract generalized features directly from raw data, capturing subtle behavioral cues such as facial expressions, speech intonation, and movement patterns. Their ability to model temporal dynamics and integrate multimodal inputs allows them to analyze complex human interactions, such as behavioral synchrony, emotion recognition, and gaze or gesture analysis, with minimal fine-tuning. Your thesis will focus on using some of the state-of-the-art foundational models to develop and evaluate a method for analyzing behavior from video and audio data.
  • Extract multimodal features using foundation models.
  • Develop or adapt synchrony metrics for these features.
  • Compare the approach with traditional methods (e.g., Motion Energy Analysis).
  • Evaluate generalizability across datasets.
Real-Time Listener Behavior Generation
Bachelor/Master
This thesis explores generating real-time listener behaviors for Epic Games' MetaHuman framework within Unreal Engine. The goal is to design and implement an interactive system that reacts dynamically to conversational inputs. The project involves integrating generative AI techniques with Unreal Engine’s MetaHuman to create realistic, responsive listener behaviors. Familiarity with Unreal Engine, MetaHuman, and generative AI models is an advantage but not mandatory.
 
 

Supervised Theses

  • Automated ICEP-R Annotation of Infant-Caregiver Interactions Using V-Jepa Self-Supervised Learning (2024, Ahmed)
  • Augmenting Social Interactive Agents: Integrating Long-Term Memory in Large Language Models (2024, Lama)
  • Interactive Agent Realism: Mediapipe 3D Blendshapes for Low Resource-Intensive Listening Behavior Modeling (2024, Sarah)

Links

Publikationen
Networking

Projects

DEEP: Mehrschichtige Verarbeitung von Emotionen für Soziale Agenten Kombination einer Interpretation von Sozialen Signalen und einem Computermodell für Emotionen von Dialogpartnern
Auswirkungen der Covid-19-Pandemie auf Elternschaft und kindliche Entwicklung (SCHWAN)

Search