H. Kasaei

Papers

Total Citations

H-Index

About

H. Kasaei is a leading researcher in open-ended robotic perception, with a focus on enabling robots to learn and recognize objects continuously in real-world, dynamic environments. Their key research areas include hierarchical Bayesian modeling, 3D object recognition, and grounding language in perception for autonomous systems. Kasaei’s most impactful contribution is the development of the Local Hierarchical Dirichlet Process (Local-HDP), a non-parametric Bayesian approach introduced in their 2021 paper (12 citations). This method allows robots to incrementally learn independent topics for each object category, adapting to new environments without forgetting prior knowledge—a critical advance for lifelong learning in robotics. Their 2014 work on grounding language in perception (10 citations) further bridges human supervision and autonomous scene conceptualization, enabling robots to interpret verbal and gestural cues. Most recently, their 2023 paper (3 citations) tackles the challenge of occluded 3D object recognition, extending Local-HDP’s robustness. With a growing citation record, Kasaei’s work is foundational for developing robots that can learn from and interact with humans in open-ended, cluttered settings, making significant strides toward truly autonomous, adaptive robotic agents.