Payam Jome Yazdian
Papers
2
Total Citations
37
H-Index
2
About
Payam Jome Yazdian is a rising researcher at the intersection of human-computer interaction, computer vision, and natural language processing, with a focus on understanding and generating expressive human motion. His work centers on bridging the gap between non-verbal communication and machine learning, particularly in the domains of co-speech gesture generation and motion description. In his highly cited 2022 paper, "Gesture2Vec," Yazdian introduced a novel representation learning method to cluster and generate co-speech gestures, addressing a critical challenge in creating more natural interactions for virtual agents and robots. This work, which has garnered 34 citations, provides a foundation for machines to better understand the nuanced relationship between speech and body language. More recently, Yazdian has pushed the boundaries of motion understanding with "MotionScript," a 2025 framework that generates fine-grained, natural language descriptions of 3D human motions. This work moves beyond simplistic action labels, offering a structured vocabulary to capture the full expressiveness of human movement. Through his innovative approaches, Yazdian is making significant contributions to embodied AI, promising more intuitive and lifelike human-agent interactions.
Research Focus
Key Achievements
Top Papers
- 1
- 2