|
Tathagata Bandyopadhyay
Researcher | Software Engineer | MSc. Informatics (TUM)
I am a research assistant in the Autonomous Vision Group (AVG) led by Prof. Dr. Andreas Geiger at Tübingen AI Center, University of Tübingen, Germany. I am currently exploring the underlying dynamics of the visual generative (diffusion / flow matching) models with a focus on understanding how the visual semantic concepts emerge, evolve and interact throughout the denoising trajectory.
📚 Additionally, I work as a part-time research engineer at Scholar-Inbox, a free comunity driven and AI powered personalized research paper recommender system developed by the researchers for the researchers.
👨🏻🎓 I completed M.Sc. in Informatics with distinction from Technical University of Munich (TUM), Germany with a focus on Deep Learning in Visual Computing. I was advised by Prof. Dr. Matthias Niessner from Visual Computing and AI Lab for my master's thesis on 3D Neural Parametric Head Models.
🤖 I love building deep learning solutions with software engineering principles. I have been fortunate to work on a variety of research projects over the years. These days, I am primarily focused on research at the overlap of:
- Diffusion / Flow Matching Denoising Dynamics
- Neural Implicit Parametric Models
- Feed-Forward Gaussian Splatting
- Representation Learning for World Models
- Latent Visual / Multi-modal Reasoning
📻 In my spare time, I occasionally experiment with radio communication and off-grid decentralized mesh networking systems, mostly out of curiosity and for the fun of understanding how they work.
Email /
CV /
MSc. Thesis /
Google Scholar /
Linkedin /
Github
|