Harkirat Behl

Papers

Total Citations

H-Index

About

Harkirat Behl is a researcher whose work lies at the intersection of computer vision, video understanding, and online learning systems. His key research areas include human action detection, video object segmentation, and meta-learning for efficient visual recognition. Behl’s major contribution is the development of incremental tube construction for human action detection—a method that enables real-time, online action detection in video streams, overcoming the limitations of traditional batch-processing systems. This work, cited 12 times, is particularly impactful for applications like human-robot interaction, where immediate response is critical. He also advanced video object segmentation through meta-learning deep visual words, allowing models to quickly adapt to novel object classes with minimal data—a crucial capability for personal robots and autonomous vehicles. This paper has garnered 9 citations. Behl’s research demonstrates a clear focus on bridging the gap between offline accuracy and online efficiency, making his contributions highly relevant for real-world deployment. His work is notable for addressing practical constraints in dynamic environments, positioning him as a promising voice in the field of video understanding and adaptive vision systems.