Publications
For a complete list of my publications, please visit my google scholar!
Papers (Conferences/Journals)
The Surprising Utility of Group Partitioning in Improving Conformal Prediction of Visual Classifiers under Distributional Shifts
K. Thopalli, V. Narayanaswamy, J. J. Thiagarajan
CVPR Uncertainty Quantification for Computer Vision Workshop 2025 [paper] Oral
Sequentially Acquiring Concept Knowledge to Guide Continual Learning
S. Kundargi, K. Thopalli, T. Gokhale
Second Workshop on Visual Concepts, CVPR 2025 [paper]
Leveraging Registers in Vision Transformers for Robust Adaptation
S. Yellapragada, K. Thopalli, V. Narayanaswamy, W. Sakla, Y. Liu, Y. Mubarka, D. Samaras, J. J. Thiagarajan
IEEE ICASSP 2025 [paper] Oral
On The Role of Prompt Construction In Enhancing Efficacy and Efficiency of LLM-Based Tabular Data Generation
B. Banday*, K. Thopalli*, T. Islam, J. J. Thiagarajan
IEEE ICASSP 2025 [paper] (* equal contribution)
Physics-informed transformation toward improving the machine-learned NLTE models of ICF simulations
M. S. Cho, P. E. Grabowski, K. Thopalli, T. S. Jayram, M. J. Barrow, J. J. Thiagarajan, R. Anirudh, H. P. Le, H. A. Scott, J. B. Kallman, B. C. Stephens, M. E. Foord, J. A. Gaffney, P.-T. Bremer
Physical Review Research 2025 [paper]
DECIDER: Leveraging Vision-Language Priors for Improved Model Failure Detection and Explanation
R. Subramanyam*, K. Thopalli*, V. Narayanaswamy*, J. J. Thiagarajan
ECCV 2024 [paper] [project page] [code] (* equal contribution)
On the Use of Anchoring for Training Vision Models
V. Narayanaswamy, K. Thopalli, R. Anirudh, Y. Mubarka, W. Sakla, J. J. Thiagarajan
NeurIPS 2024 [paper] [project page] [code] Spotlight
Speeding Up Image Classifiers with Little Companions
Y. Liu, K. Thopalli, J. J. Thiagarajan
[preprint]
Improving object detectors by exploiting bounding boxes for augmentation design
S. Devi, K. Thopalli, R. Dayana, P. Malarvezhi, J.J. Thiagarajan
IEEE Access [paper]
Advances in Computer Vision for Home-Based Stroke Rehabilitationby exploiting bounding boxes for augmentation design
K. Thopalli, N. Meniconi, T. Ahmed, S.K. Yeshala, A. Kelliher, T. Rikakis, P. Turaga
Computer Vision, pp. 109–127, Chapman and Hall/CRC (2024) [book chapter]
InterAug: A Tuning-Free Augmentation Policy for Data-Efficient and Robust Object Detection
K. Thopalli, S. Devi, J. J. Thiagarajan
ICCV 2023 Workshop on Visual Inductive Priors for Data-Efficient Deep Learning [paper] Oral
The Surprising Effectiveness of Deep Orthogonal Procrustes Alignment in Unsupervised Domain Adaptation
K. Thopalli, R. Anirudh, P.Turaga, J. J. Thiagarajan
IEEE Access 2023 [paper]
Improving Single-Stage Object Detectors for Night-Time Pedestrian Detection
S. Devi, K. Thopalli, P. Malarvezhi, J. J. Thiagarajan
International Journal on Pattern Recognition and Aritifical Intelligence 2022 [paper]
A Hierarchical Bayesian Model for Cyber-Human Assessment of Rehabilitation Movement
T. Ahmed, T. Rikakis, S. Zilevu, A. Kelliher, K. Thopalli, P. Turaga, S.L. Wolf
[preprint]
Geometric Alignment Improves Test-Time Adaptation
K. Thopalli, P. Turaga, J. J. Thiagarajan
ICML2022 Updatable Machine Learning Workshop Oral
Training Calibration-based Counterfactual Explainers for Deep Learning Models in Medical Image Analysis
J. J. Thiagarajan, K. Thopalli, D. Rajan, P. Turaga
Scientific Reports 2022 [paper]
Uncertainty-Driven Counterfactual Explainers for CXR-Based Diagnosis Models
K. Thopalli, D. Rajan, P. Turaga, J. J. Thiagarajan
ICML2022 Interpretable AI in Healthcare Workshop
Multi-Domain Ensembles for Domain Generalization
K. Thopalli, S. Katoch, P. Turaga, A. Spanias, J. J. Thiagarajan
Neurips 2021 Workshop on Distribution Shifts [poster]
Automated Movement Assessment in Stroke Rehabilitation
T. Ahmed*, K. Thopalli*, T. Rikakis, P. Turaga, A. Kelliher, J.-B. Huang, S.L. Wolf
Frontiers in Neurology, 2021 [paper] (* equal contribution)
Calibrate and Prune: Using Prediction Calibration to Improve Lottery Tickets under Distribution Shifts
B. Venkatesh, J. J. Thiagarajan, K. Thopalli, P. Sattigeri
[preprint]
Invenio: Discovering Hidden Relationships Between Tasks/Domains Using Structured Meta Learning
S. Katoch*, K. Thopalli*, J.J. Thiagarajan, P. Turaga, A. Spanias
[preprint] (* equal contirbution)
MaAST: Map Attention with Semantic Transformers for Efficient Visual Navigation
Z. Seymour, K. Thopalli, N. Mithun, H.-P. Chiu, S. Samarasekera, R. Kumar
ICRA 2020 [paper]
Perturbation Robust Representations of Topological Persistence Diagrams
A. Som*, K. Thopalli*, K.N. Ramamurthy, V. Venkataraman, A. Shukla, P. Turaga
ECCV, 2018 [paper] (* equal contibution)