Sanghyun Son

CS PhD @ UMD

Home

Curriculum Vitae

Hi, I'm Sanghyun.

I am a 4th year CS PhD student at University of Maryland. I am advised by professor Ming C. Lin at GAMMA Lab, and my research interest mainly lies in the intersection between geometry, physics, and reinforcement learning. In the last two summers, I spent wonderful time at Adobe Research, working on projects about differentiable mesh. Currently, I'm working on extending this differentiable mesh to solve various problems related to geometry, such as 3D reconstruction or generative models :)

Contact

Sanghyun Son

E-mail: [email protected]

Curriculum vitae

Publications

DMesh: A Differentiable Representation for General Meshes

Sanghyun Son, Matheus Gadelha, Yang Zhou, Zexiang Xu, Ming C Lin, Yi Zhou

Accepted to NeurIPS 2024

Arxiv Github Page

Deep Stochastic Kinematic Models for Probabilistic Motion Forecasting in Traffic

Laura Zheng, Sanghyun Son, Jing Liang, Xijun Wang, Brian Clipp, Ming C Lin

Accepted to IROS 2024

Arxiv

Visual, Spatial, Geometric-Preserved Place Recognition for Cross-View and Cross-Modal Collaborative Perception

Peng Gao, Jing Liang, Yu Shen, Sanghyun Son, Ming C Lin

IROS (Best Paper Award), 2023

Gradient Informed Proximal Policy Optimization

Sanghyun Son, Laura Yu Zheng, Ryan Sullivan, Yi-Ling Qiao, Ming Lin

NeurIPS, 2023

Arxiv Github

Traffic-Aware Autonomous Driving with Differentiable Traffic Simulation

Laura Zheng, Sanghyun Son, Ming C Lin

ICRA, 2023

Arxiv

Differentiable Hybrid Traffic Simulation

Sanghyun Son, Yi-Ling Qiao, Jason Sewall, Ming C Lin

SIGGRAPH Asia, ACM Transactions on Graphics (TOG), 2022

Arxiv Github Page

Precise Hausdorff distance computation for freeform surfaces based on computations with osculating toroidal patches

Sanghyun Son, Myung-Soo Kim, Gershon Elber

Computer Aided Geometric Design, 2021

Github

Surface--surface-intersection computation using a bounding volume hierarchy with osculating toroidal patches in the leaf nodes

Youngjin Park, Sanghyun Son, Myung-Soo Kim, Gershon Elber

Computer-Aided Design, 2020

Efficient minimum distance computation for solids of revolution

Sanghyun Son, Seunghyun Yoon, Myung-Soo Kim, Gershon Elber

Eurographics (Computer Graphics Forum), 2020

Github

Revisiting parameter sharing in multi-agent deep reinforcement learning

Justin K Terry, Nathaniel Grammel, Sanghyun Son, Benjamin Black, Aakriti Agrawal

arXiv preprint arXiv:2005.13625, 2020

Arxiv

Before joining UMD...

I have done my undergrad and Master's degree at the Seoul National University in Korea. In those years, I was advised by professor Myung-Soo Kim, and mainly worked on geometric algorithms that can be used for distance and collision queries between parametric freeform surfaces.

Share