Publications
Conferences
Samrudhdhi B Rangrej, Kevin J Liang, Tal Hassner, James J Clark. GliTr: Glimpse Transformers with Spatiotemporal Consistency for Online Action Prediction, Winter Conference on Applications of Computer Vision (WACV) 2023.
Samrudhdhi B Rangrej, Chetan L Srinidhi, James J Clark. Consistency driven Sequential Transformers Attention Model for Partially Observable Scenes, Computer Vision and Pattern Recognition (CVPR) 2022.
Kevin J Liang, Samrudhdhi B Rangrej, Vladan Petrovic, Tal Hassner. Few-shot Learning with Noisy Labels, Computer Vision and Pattern Recognition (CVPR) 2022.
Samrudhdhi B Rangrej, James J Clark. A Probabilistic Hard Attention Model For Sequentially Observed Scenes, British Machine Vision Conference (BMVC) 2021.
Karthik Gopinath*, Samrudhdhi B Rangrej*, Jayanthi Sivaswamy. A deep learning framework for segmentation of retinal layers from OCT images, Asian Conference on Pattern Recognition (ACPR) 2017. (* equal contribution)
Samrudhdhi B Rangrej, Jayanthi Sivaswamy. A biologically inspired saliency model for color fundus images, Indian Conference on Computer Vision, Graphics and Image Processing (ICVGIP) 2016.
Journals
Samrudhdhi B Rangrej, Jayanthi Sivaswamy, Priyanka Srivastava. Scan, dwell, decide: Strategies for detecting abnormalities in diabetic retinopathy, (PLoS one) 2018.
Samrudhdhi B Rangrej, Jayanthi Sivaswamy. Assistive lesion-emphasis system: an assistive system for fundus image readers, Journal of Medical Imaging (JMI) 2017.
Workshops
- Zahra Vaseqi, Ibtihel Amara, Samrudhdhi B Rangrej. Label Noise Resiliency with Self-supervised Representations, Self-Supervised Learning Theory and Practice (NeurIPS Workshop) 2021.
Pre-prints
Samrudhdhi B Rangrej, Kevin J Liang, Xi Yin, Guan Pang, Theofanis Karaletsos, Lior Wolf, Tal Hassner. Revisiting Linear Decision Boundaries for Few-Shot Learning with Transformer Hypernetworks, 2022.
Samrudhdhi B Rangrej, James J Clark. Visual Attention in Imaginative Agents, arXiv preprint 2021.
Theses
- Samrudhdhi B Rangrej. Visual perception based assistance for fundus image readers, 2017.