Summary
Sung-heon Park is a Senior Machine Learning Engineer with nine years of experience specializing in 3D vision and neural rendering, currently building neural rendering–based 3D reconstruction for autonomous driving at Lucid Motors. He holds a PhD from Seoul National University and has led research at Samsung Advanced Institute of Technology and Northwestern, delivering fast-training dynamic NeRFs, few-shot NeRF methods, and dynamic 3D Gaussian splatting that couples instance segmentation with reconstruction. His work has practical impact in consumer products (face liveness and white balance improvements integrated into Samsung devices) and in academic systems for photorealistic, pose-drivable human avatars from monocular video. Recognized internally at Samsung with top awards, he blends rigorous academic methods with production-minded engineering, optimizing for model size, speed, and accuracy. Based in Newark, California, he brings both deep research credentials and hands-on system-building to bridge 4D scene understanding and real-world deployment. An underappreciated strength is his focus on test-time refinement and low-rank optimization to make advanced neural rendering techniques tractable for real-time applications.
9 years of coding experience
6 years of employment as a software developer
Doctor of Philosophy - PhD Intelligent Systems, Doctor of Philosophy - PhD Intelligent Systems at Seoul National University
Bachelor's degree Exchange Student, Bachelor's degree Exchange Student at RMIT University
Master's degree Computer Science, Master's degree Computer Science at Korea Advanced Institute of Science and Technology