Junhwa Hur
Research Scientist at Google
Cambridge, Massachusetts, United States
Summary
👤
Senior🎓
Top SchoolJunhwa Hur is a research scientist based in Cambridge, MA with nine years of experience at the intersection of computer vision and machine learning. Currently at Google, Junhwa applies a strong academic foundation—a PhD from Technische Universität Darmstadt and an MS from Seoul National University—to research problems in 3D dynamic scene understanding, multi-task and self-supervised learning. His background includes developing RGB-D pipelines for deformable object modeling and hands-on work in motion, depth, occlusion, and semantic segmentation from graduate research to industry internships. Junhwa’s trajectory shows a rare blend of deep theoretical training and practical system-building across academia and industry. He maintains a personal webpage for his work and has steadily progressed from early internships to influential research roles, signaling both technical depth and adaptability. Colleagues value him for translating complex perception problems into reproducible experimental systems.
9 years of coding experience
5 years of employment as a software developer
Master of Science - MS, Electrical and Computer Engineering, Master of Science - MS, Electrical and Computer Engineering at Seoul National UniversityBachelor of Science - BS, Electronics and Electrical Engineering, Bachelor of Science - BS, Electronics and Electrical Engineering at Pohang University of Science and TechnologyTechnischen Universität Darmstadt
Github Skills (20)
depth-estimation10
ieee10
occlusion10
optical-flow10
scene10
computer-vision9
autonomous-driving8
deep-learning8
pytorch8
siggraph7
autonomous-vehicles7
backpropagation5
procedural-generation5
neural-rendering5
machine-learning4
Programming languages (2)
Github contributions (5)
visinf/irr
Jun 2019 - May 2021
Iterative Residual Refinement for Joint Optical Flow and Occlusion Estimation (CVPR 2019)
Contributions:24 commits, 1 PR, 22 pushes in 1 year 11 months
occlusionoptical-flow
hurjunhwa/mld_crf
Feb 2017 - Apr 2020
Multi-lane detection in urban driving environments using conditional random fields
Contributions:8 commits, 1 PR, 7 pushes in 3 years 2 months
lane-detectionrandom-fieldsconditionallanefields
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.