Peiyu Zeng
Papers
1
Total Citations
2
H-Index
1
About
Peiyu Zeng is a researcher at the forefront of computer vision and robotics, with a specialized focus on bridging the critical gap between simulation and real-world perception. His primary research areas include physics-grounded sensor simulation, active stereo vision, and domain adaptation for autonomous systems. Zeng’s most notable contribution is his pioneering work on "Close the Optical Sensing Domain Gap by Physics-Grounded Active Stereo Sensor Simulation," which introduces a fully physics-grounded pipeline for simulating active stereovision depth sensors. By incorporating material acquisition and ray-tracing techniques, his approach enables highly realistic sensor outputs, significantly reducing the sim-to-real domain gap that hampers the deployment of learned models in real-world applications. This work has garnered attention in both academic and industry communities, laying a foundation for more robust robotic perception systems. Zeng’s research is instrumental in advancing the reliability of autonomous systems, making him a rising voice in the field of embodied AI and simulation-based learning.
Research Focus
Key Achievements
Top Papers
- 1