Email
Scholar
LinkedIn
Researcher • Developer • Scientist
Geometric generative models, multimodal image processing, and remote sensing
I’m a first‑year Computer Science PhD student at IISc, advised by Prof. Chiranjib Bhattacharya. I also work as a Scientist/Engineer‑SD at ISRO (SAC). My current research focuses on theoretically analysing the geometric representation of generative models and applications to multi‑modal image handling. Previously, I’ve worked on model stability, robustness and convergence.
Updates
- Oct 2024 : Paper F2former: When Fractional Fourier Meets Deep Wiener Deconvolution and Selective Frequency Transformer for Image Deblurring is accepted at WACV, 2025.
- Sep 2024 : Preprint: F2former: When Fractional Fourier Meets Deep Wiener Deconvolution and Selective Frequency Transformer for Image Deblurring.
- Aug 2024: One paper A Sinkhorn Regularized Adversarial Network for Image Guided DEM Super-resolution using Frequency Selective Hybrid Graph Transformer is accepted in ICPR 2024.
- Jul 2024: One paper A Framework for Realistic Paired Dataset Generation for Deep Learning Based Restoration of Satellite Images published in IEEE IGARSS 2024.
- May 2024: One paper C-LiSA: A Hierarchical Hybrid Transformer Model Using Orthogonal Cross Attention for Satellite Image Cloud Segmentation published in IEEE TGRS.
- Mar 2024: One paper: High Resolution Multi-Spectral Image Guided DEM Super-resolution using Sinkhorn Regularized Adversarial Network is accepted in ICLR ML4RS 2024.
- Nov 2023: Preprint: CLiSA: A Hierarchical Hybrid Transformer Model using Orthogonal Cross Attention for Satellite Image Cloud Segmentation
- Jul 2022: Published paper: Hiding images within audio using deep generative model at Multimedia Tools and Applications journal.
Reviewer Service
IEEE TGRS
CVPR 2024
ECCV 2024
ICPR 2024
WACV 2025
Selected Publications
Sinkhorn Regularized Adversarial Network for Image‑Guided DEM Super‑Resolution using Frequency‑Selective Hybrid Graph Transformer
A framework for Realistic Paired Dataset Generation for Deep Learning based Restoration of Satellite Images
Hiding Images within Audio using Deep Generative Model
Experience
-
Scientist/Engineer‑SD — Space Applications Centre (ISRO) • Jan 2021 – Present
- Operationalised deep band reconstruction & restoration pipelines; striping correction; semantic mapping for high‑res satellite imagery (deployed at NRSC).
- Mission work: OM for HRSAT; PM for Cartosat-3S.
- Research: DEM SR (Sinkhorn + Graph Transformer), Fractional‑Fourier deblurring transformer; Lipschitz‑validated segmentation.
-
PhD (Computer Science) — Indian Institute of Science (IISc) • 2025 –
- Research focus: geometric representation of generative models; multimodal image handling.