Tathagata Bandyopadhyay

I love building deep learning solutions with software engineering principles, and I am currently working on multi-view 3D reconstruction with sparse feed-forward Gaussian Splatting as part of my research internship in Autonomus Learning Group led by Prof. Dr. Georg Martius, at Tübingen AI Center, University of Tübingen, Germany.

Previously, I worked at Siemens (Technology R&D) to build a RAG based query framework for large industrial knowledge graphs using LLMs and LangChain followed by a research internship at the Max Planck Institute for Intelligent Systems, Tübingen, where I developed a video quality analyzer tool using OpenCV, ffmpeg, YOLO and vision language models like CLIP.

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. Some of my other deep learning projects involves semi-supervised learning, knowledge distillation, target speaker extraction, music retrieval or cover song identification, neural object relighting, and so on using Python and PyTorch.

Topics of interest:
Multi-modal understanding, latent representation learning, hyperbolic deep learning, 3D reconstruction, sparse gaussian splatting, autoencoders, diffusion models, GANs, large language models, transformers, topological deep learning, brain computer interface.

Email  /  CV  /  MSc. Thesis  /  Google Scholar  /  Linkedin  /  Github

Technical University of Munich, Germany

Kalinga Institute of Industrial Technology, Bhubaneswar, India

University of Tübingen, Germany

Max Planck Institute for Intelligent Systems, Tübingen, Germany

Siemens (Technology R&D), Munich, Germany

Dell Technologies, Bangalore, India

Website design and source code adapted from Jon Barron's website.