Po-Jen Lai

Papers

Total Citations

H-Index

About

Po-Jen Lai is a researcher advancing the intersection of computer vision and robotics, with a primary focus on learning collaborative and interactive behaviors from unstructured video data. His most-cited work, "Learning Collaborative Action Plans from YouTube Videos" (2022), introduces a novel framework for extracting and modeling multi-agent coordination directly from real-world video demonstrations. By leveraging large-scale, in-the-wild footage, Lai’s approach enables robots to infer structured action plans for human-robot collaboration without requiring manual annotations or simulated environments. This contribution addresses a critical bottleneck in robotic learning—the scarcity of labeled interaction data—and has already garnered 5 citations, signaling growing interest from the community. Beyond this paper, Lai’s research spans action recognition, imitation learning, and video understanding, with an emphasis on making robots more capable of interpreting and participating in complex social tasks. His work is notable for bridging the gap between raw visual observations and actionable robotic policies, offering a scalable path toward more autonomous and socially aware systems. For students and researchers, Lai’s research represents a compelling example of how computer vision can directly empower robotics, turning passive video watching into active, collaborative intelligence.