I am an Associate Professor in the Department of Computer Science and Software Engineering at Auburn University. My research focuses on open-world learning under limited supervision, with an emphasis on how prior knowledge, structure, and interaction can enable robust perception and reasoning. My work spans computer vision, neuro-symbolic AI, embodied and active perception, event understanding, and applications in data-scarce domains such as genomics, agriculture, robotics, and security. I am a recipient of the NSF CAREER Award and currently lead projects on multimodal event understanding, active perception, and open-world AI. Prior to joining Auburn, I was an Assistant Professor in the Department of Computer Science at Oklahoma State University, Stillwater.
I received my Ph.D. in Computer Science and Engineering from the University of South Florida, where I was fortunate to be advised vy Dr. Sudeep Sarkar in the Computer Vision and Pattern Recognition Group. I received my Master's degree in Management Information Systems from the Muma College of Business at the University of South Florida and my undergraduate degree in Electronics and Communication Engineering from Velammal Engineering College, Anna University, India.
My research lies at the intersection of computer vision, machine learning, and artificial intelligence, with a focus on building systems that can understand, reason, and act in open-world environments. A central theme of my work is moving beyond static recognition toward structured, adaptive visual intelligence: systems that can infer events, relationships, affordances, and intent from limited supervision and incomplete observations. To this end, my group develops models that combine representation learning with prior knowledge, causal and probabilistic reasoning, neuro-symbolic structure, and interaction. Recent projects include event-centric video understanding, embodied active perception, visual relationship reasoning, commonsense-guided grounding, open-world scene understanding, and multimodal learning for data-scarce domains. We evaluate these ideas across applications in robotics, genomics, agriculture, biomedical AI, manufacturing, and security, where robust generalization under uncertainty is essential.