Sheela Kurupathi

Sheela Raju Kurupathi is research associate at the Interactive and Cognitive Systems (ICS) group  in the Vision and Imaging Technologies department .

Her R&D activities are related to the following areas such as machine learning, deep learning, artificial intelligence, computer vision, and human-computer interaction.

For further details about her activities please check the subsections below.


Sheela Raju Kurupathi is a deep learning researcher. Her research is mainly focused on the development of machine learning and AI based systems for human-computer interaction.

Sheela received a Master’s degree in Informatik from Technical University Kaiserslauern with high distinction in 2019. She received scholarship awards like Deutschland Stipendium and DAAD Stibet during the course of her Masters for her outsatanding performance. She worked in various research institutes like DFKI, Fraunhofer ITWM, SAP as a research assistant and gained experience in the field of AI. She also worked as a teaching assistant in the Technical University Kaiserslauern during her Masters. 

She has done her guided research on "Restoring document degradations sing conditional GANs" in DFKI and later worked on her master's thesis in "Generating humans based on pose and clothing using GANs". Later she joined as a full time researcher in DFKI in the area of deep learning wih focus on GAN's. 

She gained practical experience by working in various areas like deep learning, machine learning, federated learning, and computer vision. Currently her main research is focused on interactive machine learning in the domain of cognitive systems.



Kurupathi, Sheela. (2020). "Survey on Federated Learning Towards Privacy Preserving AI",  CSEIT 2020, Denmark.


Kurupathi, Sheela & Murthy, Pramod & Stricker, Didier. (2020). "Generation of Human Images with Clothing using Advanced Conditional Generative Adversarial Networks", DeLTA 2020, France.



Kurupathi, Sheela & Dumpala, Veeru & Bukhari, Syed & Dengel, Andreas. (2019). "Removal of Historical Document Degradations using Conditional GANs", ICPRAM 2019, Prague.