Summary
Bo-hsun Chen is a Computer Science PhD candidate at the University of Wisconsin–Madison specializing in camera sensor simulation, photorealistic rendering, and sim-to-real transfer for robotic perception. He builds physics-informed, differentiable camera models and integrates ML methods (NeRFs, GANs, PyTorch/CUDA) to reduce reliance on real imagery—his work cut required real photos by over 80% while maintaining performance. He led creation of the POLAR-Sim dataset that augments NASA data with 3D meshes and dense labels for detection, segmentation, and stereo tasks, and has parallelized BRDF optimization in CUDA for scalable inverse rendering. With a strong interdisciplinary background spanning mechanical, electrical, and computer engineering, he combines hands-on GPU programming and robotics experimentation with rigorous ML evaluation. Open to internships in 2026, he focuses on simulation-driven AI that accelerates robotics research and lowers development costs.
9 years of coding experience
4 years of employment as a software developer
Doctor of Philosophy - PhD, Computer Science, 3.89/4.0, Doctor of Philosophy - PhD, Computer Science, 3.89/4.0 at University of Wisconsin-Madison
Master of Science - MS, System Control Divion, Mechanical Engineering, 4.05/4.3, Master of Science - MS, System Control Divion, Mechanical Engineering, 4.05/4.3 at National Taiwan University
Chinese, English, Japanese