Suk‐Hwan Lee

About

Suk-Hwan Lee is a researcher at the forefront of computer vision and deep learning, with a specialized focus on 3D object analysis and visual tracking. His work bridges the gap between theoretical deep learning models and practical applications in autonomous systems, robotics, and augmented reality. Lee’s most impactful contribution is his development of novel deep learning architectures for 3D point cloud classification, notably GSV-NET, a multi-modal network that enhances LiDAR-based perception for autonomous vehicles and robotics. He has also pioneered methods for 3D object classification and retrieval using advanced geometric signatures like the Wave Kernel Signature and Global Point Signature Plus, integrated with deep wide residual networks. In visual tracking, Lee has applied deep reinforcement learning, specifically DQN agents, to improve object tracking in virtual environmental simulations. With over 60 citations across his top papers, his work is gaining recognition for advancing the robustness and efficiency of 3D computer vision systems. Lee’s research is particularly notable for its practical orientation, directly addressing challenges in smart cars, robot navigation, and multimedia content processing, making him a rising contributor to the field of intelligent visual perception.

Research Focus

Key Achievements

Top Papers

1. 1
   Deep Reinforcement Learning-Based DQN Agent Algorithm for Visual Object Tracking in a Virtual Environmental Simulation

   30 citations · 2022
2. 2
   GSV-NET: A Multi-Modal Deep Learning Network for 3D Point Cloud Classification

   16 citations · 2022
3. 3
   A Deep Learning Method for 3D Object Classification and Retrieval Using the Global Point Signature Plus and Deep Wide Residual Network

   11 citations · 2021
4. 4
   A Deep Learning Method for 3D Object Classification Using the Wave Kernel Signature and A Center Point of the 3D-Triangle Mesh

   7 citations · 2019

Key Collaborators

Contact & Links