Pradeep Yarlagadda

International Institute of Information Technology

Papers

Total Citations

H-Index

About

Pradeep Yarlagadda’s research lies at the intersection of computer vision and cognitive science, with a primary focus on computational models of visual attention—specifically, saliency prediction. His work seeks to bridge the gap between human visual cognition and machine perception, aiming to make robots and AI systems more adept at understanding what humans find visually interesting. His most cited paper, “Tidying Deep Saliency Prediction Architectures” (2020), critically examines the data-driven deep learning approaches that have come to dominate saliency estimation. Rather than simply proposing a new model, Yarlagadda’s contribution is a thoughtful analysis and refinement of existing architectures, helping to streamline and improve the performance of neural networks for this task. While his citation count is still growing, his work is foundational for researchers looking to understand the structural choices behind effective saliency models. Yarlagadda’s research is particularly valuable for students and engineers working on human-robot interaction, autonomous driving, and image compression, where knowing where humans look is key. His approach reflects a deep commitment to making machine vision not just more accurate, but more human-like.