A parts-based occlusion-aware network for robust person re-identification that addresses partial visibility challenges through structured part-level feature learning.
@article{dutta2025poanet,title={{POANet}: Parts-based Occlusion Aware Network for Person Re-{ID}},author={Dutta, Pratyay and Jonnalagedda, Padmaja and Bhanu, Bir},journal={IEEE Transactions on Biometrics, Behavior, and Identity Science (TBIOM)},year={2025},}
TPAMI, Under Review
SAKE: Structure Aware Kernel Experts for Event-Based Object Detection
Pratyay
Dutta, Ankith Jain
Rakesh Kumar, Nghia
Nguyen, and
1 more author
In IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2025
SAKE replaces costly Transformer self-attention with a physics-inspired heat-diffusion mechanism that propagates information along object edges to improve localization. By using structure-aware kernels and frequency-domain routing, it achieves high-performance event-based detection with significantly lower computational overhead.
@inproceedings{dutta2025heatdiffusion,title={SAKE: Structure Aware Kernel Experts for Event-Based Object Detection},author={Dutta, Pratyay and Rakesh Kumar, Ankith Jain and Nguyen, Nghia and Bhanu, Bir},booktitle={IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)},year={2025},}
TCSVT
Latent Diffusion-Guided Feature Inpainting for Occluded Person Re-Identification With Hybrid Re-Ranking
Pratyay
Dutta and Bir
Bhanu
IEEE Transactions on Circuits and Systems for Video Technology (TCSVT), 2025
A Latent Diffusion-guided De-Occluder (DDO) that inpaints corrupted features to reconstruct clean, identity-preserving representations for more robust person re-identification. Combined with a Hybrid Re-Ranking (HRR) scheme, this approach eliminates the need for explicit occlusion modeling and achieves new state-of-the-art results.
@article{dutta2025latentdiffusion,title={Latent Diffusion-Guided Feature Inpainting for Occluded Person Re-Identification With Hybrid Re-Ranking},author={Dutta, Pratyay and Bhanu, Bir},journal={IEEE Transactions on Circuits and Systems for Video Technology (TCSVT)},year={2025},note={Minor revision undergoing},}
2022
TCBB
Hierarchical Vaccine Allocation based on Epidemiological and Behavioral Considerations
Satyaki
Roy, Pratyay
Dutta, and Preetam
Ghosh
IEEE/ACM Transactions on Computational Biology and Bioinformatics (TCBB), 2022
An optimized hierarchical vaccine distribution strategy incorporating epidemiological and behavioral factors with clustering-based zone allocation for improved vaccine redistribution across demographics.
@article{roy2022hierarchical,title={Hierarchical Vaccine Allocation based on Epidemiological and Behavioral Considerations},author={Roy, Satyaki and Dutta, Pratyay and Ghosh, Preetam},journal={IEEE/ACM Transactions on Computational Biology and Bioinformatics (TCBB)},year={2022},}
2021
BIBM
Generalizable Multi-Vaccine Distribution Strategy based on Demographic and Behavioral Heterogeneity
Satyaki
Roy, Pratyay
Dutta, and Preetam
Ghosh
In IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 2021
A generalizable multi-vaccine distribution strategy using linear optimization and clustering algorithms to address demographic and behavioral heterogeneity in the Chicago district.
@inproceedings{roy2021multivaccine,title={Generalizable Multi-Vaccine Distribution Strategy based on Demographic and Behavioral Heterogeneity},author={Roy, Satyaki and Dutta, Pratyay and Ghosh, Preetam},booktitle={IEEE International Conference on Bioinformatics and Biomedicine (BIBM)},year={2021},}
