Research

Artificial Intelligence, Machine Learning and Deep Learning with applications to intelligent systems and real-world decision-support systems.

Key Work & Outcomes

• Research includes deep learning architectures, generative models, transformers and optimized AI models.
• Research emphasizes performance-efficient architectures suitable for resource-constrained environments.

Image processing, computer vision, image restoration, deblurring and visual data enhancement.

Key Work & Outcomes

• Ph.D. thesis: Development of an Optimized System for Indoor and Outdoor Blurred Images.
• Developed a unified framework for restoring degraded visual data across image, video and text modalities.
• Proposed VAEWGAN, a hybrid generative model integrating Variational Autoencoders with Wasserstein GANs for image deblurring.
• Designed ViConNet, a spatio-temporal deep network for real-time video deblurring.
• Developed TextFormer, a transformer-based architecture for blurred document and scene-text restoration.

Signal processing techniques with applications to intelligent visual and computational systems.

Development of intelligent visual systems using deep learning and computer vision.

AI-driven decision-support systems for real-world applications.