I am currently a Postdoctoral Fellow at 3IA Côte d’Azur, working with the STARS team at Inria – Université Côte d’Azur under Prof. François Brémond. My research focuses on video understanding and reasoning, with an emphasis on anomaly detection and human-to-human interaction analysis.

  • Anomaly detection and human interaction analysis in collaboration with Toyota Research Institute (Europe & Japan).
  • Human behaviour understanding in collaboration with CobTek, Université Côte d'Azur.

Previously, I was a Project Scientist-I at the Artificial Intelligence and Robotics Lab (AIRL), Indian Institute of Science (IISc), where I also served as a Co-Principal Investigator for an NPTEL course on AI-driven perception, learning, and mapping for drones (watch on YouTube), focused on practical education in autonomous navigation and aerial robotics. I have also worked as a Consultant at Visist.AI, developing computer vision solutions for noisy, high-speed environments, particularly in sports analytics, with an emphasis on robustness and real-time performance.

I earned my Ph.D. from the Robert Bosch Centre for Cyber-Physical Systems (RBCCPS), IISc, under the supervision of Prof. Suresh Sundaram (AIRL–IISc). During my doctoral research, I also had the opportunity to work with Prof. Sundarrajan, Dr. Chandan Gautham, and Dr. Arjun Jain.

research statement

My research lies at the intersection of computer vision and robotics, with a focus on video understanding, multi-sensor fusion, continual learning, and robust generalization to unseen domains. I am particularly interested in building reliable perception systems that can operate effectively in complex, dynamic, and safety-critical real-world environments.

I have worked extensively on deploying task-specific computer vision systems for sports analytics in fast-paced and highly dynamic settings, emphasizing robustness, real-time performance, and reliable operation under challenging real-world conditions.

At the Artificial Intelligence and Robotics Lab and the Robert Bosch Centre for Cyber-Physical Systems, I contributed to multiple deployment-oriented research projects funded through academic and industrial collaborations, including initiatives associated with startups incubated at the Society for Innovation and Development and ARTPARK. My work involved the evaluation and development of state-of-the-art deep learning approaches for segmentation, object detection, and change detection across diverse robotic platforms, including ground vehicles, aerial systems, and satellite-based sensing platforms. These systems integrated multimodal sensing modalities such as RGB imagery, Infrared (IR), Side-Scan Sonar, and Synthetic Aperture Radar (SAR).

In addition, my research interests include the development of sim-to-real evaluation frameworks and the use of generative AI techniques to enhance robustness, domain adaptation, and generalization in challenging operational environments.