I am a Computer Vision and Deep Learning Researcher.
My interest lies in Multimodal Learning, 3D Human Motion Modeling, and Generation.
I have recently spent three months at Visual Computing & Artificial Intelligence group at Technical University of Munich
with Prof. Matthias Nießner.
I have completed MS by Research at CVIT, IIIT
Hyderabad under the guidance of Prof. C.V. Jawahar and Prof. Vinay P. Namboodiri. My graduate research focused on
Lip-Sync, Talking Head Generation, and Face Reenactment, along with their optimization for real-world
problems. I also have worked on the task of Table Detection in Document Images with high accuracy
under the guidance of Prof. C.V. Jawahar and Dr. Ajoy Mondal. Prior to this, I was working as a Data
Scientist with a couple of companies, broadly in the domain of Facial Recognition, Video
surveillance using AI, and Document Image Processing. I have also been actively involved with start-ups as an advisor and consultant.
My work has been published in top computer vision and machine learning conferences.
Contact: madhav
[CV] | [Google Scholar] | [LinkedIn] | [GitHub] | [Mail] | [arXiv] |
Dec 2023: Understanding the Generalization of Pretrained Diffusion Models on Out-of-Distribution Data got accepted to AAAI 2024 (Oral).
Sept 2023: Joined Visual Computing & Artificial Intelligence, Technical University of Munich as a Scientific Researcher under Prof. Matthias Nießner.
June 2023: Successfully defended MS thesis Face Reenactment: Crafting Realistic Talking Heads for Enhanced Video Communication and Beyond.
June 2023: Invited at AKG Engineering College to give Guest Lecture on Generative AI.
May 2023: Dataset Agnostic Document Object Detection got accepted to Pattern Recognition journal.
Apr 2023: Awarded with 'Non-Academic Award' by IIIT-Hyderabad for contribution towards Mental Health on campus.
Sep 2022: Compressing Video Calls using Synthetic Talking Heads got accepted to BMVC 2022.
Aug 2022: Audio-Visual Face Reenactment got accepted to WACV 2023.
Jan 2021: Joined CVIT, IIIT-Hyderabad as a full-time MS by Research student under Prof. C.V. Jawahar.
Oct 2020: CDeC-Net: Composite Deformable Cascade Network for Table Detection in Document Images got accepted to ICPR 2020 (