Soumya Ranjan Mishra

National Institute of Technology Durgapur

Papers

Total Citations

H-Index

About

Soumya Ranjan Mishra is a researcher whose work lies at the intersection of computer vision, machine learning, and human activity recognition. He is best known for developing a novel "PSO based combined kernel learning framework" for recognizing first-person (egocentric) activities in video, a contribution that addresses the challenge of understanding actions from a camera wearer’s perspective. This framework, published in 2018, has garnered 11 citations and demonstrates his ability to integrate swarm intelligence with kernel methods for robust feature learning. His research has practical implications for wearable computing, surveillance, and assistive technologies, where accurate activity recognition is critical. Mishra’s work stands out for its innovative use of particle swarm optimization to optimize kernel combinations, enhancing classification performance in complex, real-world video data. By focusing on first-person activity recognition, he has contributed to a growing field that bridges human behavior understanding and artificial intelligence. His achievements reflect a commitment to advancing machine learning techniques for video analysis, making his research valuable for students and practitioners exploring egocentric vision and intelligent systems.