Shih-Yang Su

Papers

Total Citations

H-Index

About

Shih-Yang Su is a researcher at the forefront of bridging the gap between simulated and real-world robotic control. His primary research areas include visual semantic segmentation, robot learning, and sim-to-real transfer—a critical challenge in deploying autonomous systems safely and efficiently. Su’s most influential work, “Virtual-to-Real: Learning to Control in Visual Semantic Segmentation” (2018), has garnered 69 citations, underscoring its significance in the field. This paper tackles the persistent “reality gap” between synthetic training data and real-world visual inputs, proposing a framework that allows robots to learn control policies from virtual environments without costly or hazardous physical data collection. By enabling more robust and adaptable robotic systems, Su’s contributions directly address a fundamental bottleneck in autonomous navigation and manipulation. His research not only advances the practical deployment of robots in unstructured settings but also inspires further exploration into domain adaptation and reinforcement learning. For students and researchers interested in the intersection of computer vision and robotics, Su’s work offers a compelling blueprint for leveraging simulation to achieve real-world performance.