Jelena BratulićDoctoral Researcher | PhD Student
Computer Vision Lab
Office location:
Phone: +49 761-203-8267 |
![]() |
Personal | Projects | Curriculum Vitae | Teaching and Supervision |
---|
Supervision
Please check the following guidelines if you are interested in pursuing a thesis or project with me. Here you can find details about the currently available project and some previous project. If you are interested in pursuing a project or thesis with me, please use the official LMB application process to apply for a thesis or a project.Ongoing
-
Master's Project: Improving cell segmentation and tracking via mechanistic models of rod-shaped bacteria (Moritz Steinmaier) - Supervision together with Jonas Pleyer
-
Master's Project: Automatic discovery and matching of rare skin disease wounds (Ayush Gupta) - Supervision together with Julia Hindel
-
Master's Project: Efficient VLA training (Devika Paneer Selvam)
Previous theses and projects
-
Bachelor's thesis: Exploring Visual In-Context Task Vectors (Michel Mackert, September 2025)
Teaching
Summer Semester 2024/2025
- Seminar: Block Seminar on Deep Learning - Supervisor
Paper: DeepSeek-V3 Technical Report (Student: Vladyslav Moroshan) - Seminar: Block Seminar on Deep Learning - Supervisor
Paper: Qwen2.5-VL Technical Report (Student: Amal Abed)
Winter Semester 2024/2025
- Seminar: Block Seminar on Deep Learning - Supervisor
Paper: Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution (Student: Urs Spiegelhalter)
Summer Semester 2023/2024
- Seminar: Block Seminar on Deep Learning - Supervisor
Paper: Memory Consolidation Enables Long-Context Video Understanding (Student: Michel Mackert) - Seminar: Current Works in Computer Vision - Supervisor
Paper: Sequential Modeling Enables Scalable Learning for Large Vision Models (Student: Marius Birmele)
Winter Semester 2023/2024
- Seminar: Block Seminar on Deep Learning - Supervisor
Paper: The dawn of LMMs: preliminary explorations with GPT-4V (Student: Redi Muharremi) - Seminar: Current Works in Computer Vision - Supervisor
Paper: Segment Anything (Student: Simon Baermann)