Namitha Padmanabhan

I am a first year PhD student in the Department of Computer Science at the University of Maryland (UMD), advised by Prof Abhinav Shrivastava, where I study computer vision. Prior to this, I completed my master's in Computer Science from UMD.

I received my bachelor's degree in Computer Science from RVCE Bangalore. During my junior and senior years, I had the privilege of working at IISc, with Prof Y Narahari on multi-armed bandits and Prof K V S Hari. Before pursuing my master's at UMD, I worked at Cisco for 2 years, on web microservices.

My current research interests lie in understanding implicit neural representations and exploring their utility in computer vision tasks.

Email  /  Google Scholar  /  LinkedIn

profile photo
xinc_teaser Explaining the Implicit Neural Canvas (XINC): Connecting Pixels to Neurons by Tracing their Contributions
Namitha Padmanabhan*, Matthew Gwilliam*, Pulkit Kumar, Shishira R Maiya, Max Ehrlich, Abhinav Shrivastava
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024
Project Page |Paper |Code

XINC dissects Implicit Neural Representation (INR) models to understand how neurons represent images and videos and to reveal the inner workings of INRs.

tats_framework Trajectory-aligned Space-time Tokens for Few-shot Action Recognition
Pulkit Kumar, Namitha Padmanabhan, Luke Luo, Saketh Rambhatla, Abhinav Shrivastava
Proceedings of the European Conference on Computer Vision (ECCV), 2024
Project Page |Paper

Few-shot action recognition with disentanglement of motion and appearance representations by harnessing point trackers and self-supervised representations.

diff_ssl Do Text-free Diffusion Models Learn Discriminative Visual Representations?
Matthew Gwilliam*, Soumik Mukhopadhyay*, Yosuke Yamaguchi, Vatsal Agarwal, Namitha Padmanabhan, Archana Swaminathan, Tianyi Zhou, Abhinav Shrivastava
Proceedings of the European Conference on Computer Vision (ECCV), 2024
Project Page |Paper

Exploring diffusion models as unified unsupervised image representation learning models for many recognition tasks. Proposed DifFormer and DifFeed, novel mechanisms for fusing diffusion features for image classification.

metabit Leveraging Bitstream Metadata for Fast, Accurate, Generalized Compressed Video Quality Enhancement
Max Ehrlich, Jon Barker, Namitha Padmanabhan, Larry S Davis, Andrew Tao, Bryan Catanzaro, Abhinav Shrivastava
IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2024
Paper

Restore detail in compressed videos by leveraging structure and motion information from the video bitstream, handling various compression quality settings.

preprint_diff_ssl Diffusion Models Beat GANs on Image Classification
Matthew Gwilliam*, Soumik Mukhopadhyay*, Vatsal Agarwal, Namitha Padmanabhan, Archana Swaminathan, Tianyi Zhou, Abhinav Shrivastava
preprint only
Project Page |Paper

Show the potential of diffusion models as unified unsupervised image representation learners.


Template credits