Kyungmin Lim

Research Scientist in Image/Video Processing and Multimodal AI
Principal Engineer, Samsung Research

I am a Research Scientist with over 10 years of experience in image and video processing, computer vision, and AI-driven imaging systems.

My work focuses on translating cutting-edge research into real-world consumer products, with an emphasis on perceptual image quality, real-time performance, and system-level optimization. I have contributed to award-winning products, including the CES 2024 Innovation Award.

My research interests include deep learning–based image enhancement, generative models (GANs) for super-resolution and HDR, multimodal AI such as Vision-Language Models (VLM), and perceptual image quality optimization for displays, VR, and video coding systems.

View Projects Contact
Kyungmin Lim

About

Career

Samsung Research - Principal Engineer

Mar 2015 - Present

Leading various research and development projects at Samsung Research, including Vision-Language Models, Light Field Display, AI-based image quality enhancement, VR motion sickness mitigation, and video coding standardization.

Education

Yonsei University - School of Electrical and Electronic Engineering

M.S./Ph.D. (Mar 2008 - Feb 2015)

Dissertation: Low Complexity Intra Prediction Methods for Video Coding

Yonsei University - School of Electrical and Electronic Engineering

B.S. (Mar 2004 - Feb 2008)

Technical Skills

Programming Languages
C/C++ Python Matlab
AI / Machine Learning
Deep Learning Generative Models (GANs) Super-Resolution HDR Multimodal Models Vision-Language Models
Computer Vision & Image Processing
Image/Video Processing Light Field Imaging Image Quality Enhancement Perceptual Quality Optimization

Research Projects

Major Research and Development Projects

Vision-Language Model for Desktop AI

Vision-Language Model (VLM) for Desktop AI

Developed an optimized Vision-Language Model for on-device AI applications, enabling seamless multimodal interactions on desktop platforms.

Deep Learning VLM On-Device AI
Light Field Display Research

Light Field Display Research

Designed localized 2D-to-3D conversion algorithms for stereoscopic rendering, developed crosstalk reduction algorithms based on Light Field image processing, built deep learning-based multi-layer Light Field decomposition models, and developed real-time telepresence systems for large screens.

Light Field 3D Display Deep Learning
AI for Image Quality Enhancement

AI-Driven Image Quality Enhancement

Enhanced MicroLED image quality through subpixel rendering using DNN, developed super-resolution models for upscaling SD/HD/FHD to 4K/8K on digital TVs, and developed single-image HDR enhancement technology for smartphone photography.

Super-Resolution HDR GANs Image Processing
VR Motion Sickness Mitigation

VR Motion Sickness Mitigation

Developed a system that analyzes VR content motion sickness severity in real-time using image processing techniques, dynamically adjusts field of view, and stabilizes video to reduce user discomfort.

VR Computer Vision Real-time Processing
Video Coding Standardization

Video Coding Standardization (VVC)

Contributed to the development of next-generation video compression algorithms after HEVC, improving coding efficiency.

Video Coding VVC Standardization

Awards

  • CES 2024 Innovation Award Production – Samsung 2D/3D Gaming Monitor
  • Samsung Research S/W Awards (2022) – Development of LF image generation technology that provides telepresence experience of LF Display
  • Samsung Research achievement award for R&D in the first half of the year (2021) – Volumetric Display core technology development
  • BK21 (Brain Korea 21) scholarship (Jan 2009 – Dec 2012) – National Research Foundation of Korea
  • Creative Excellence Award (Jun 2007) – Yonsei University School of Electrical and Electronic Engineering

Publications

Journals

Artifact-free crosstalk reduction in light field display based on multi-view image correlation

Kyungmin Lim, Kangwon Jeon, Jaesung Lee, Daeyeong Kim

Optica Open. Preprint, 2024

A novel method for VR sickness reduction based on dynamic field of view processing

Kyungmin Lim, Jaesung Lee, Kwanghyun Won, Nupur Kala and Tammy Lee

Virtual Reality, vol. 25, 331-340, 2021

Fast PU Skip and Split Termination Algorithm for HEVC Intra Prediction

Kyungmin Lim, Jaeho Lee, Seongwan Kim and Sangyoun Lee

IEEE Transactions on Circuits and Systems for Video Technology, vol. 25, Issue. 8, page 1335-1346, Dec, 2014

Fast Block Size and Mode Decision Algorithm for Intra Prediction in H.264/AVC

Kyungmin Lim, Seongwan Kim, Jaeho Lee, Daehyun Pak, and Sangyoun Lee

IEEE Transactions on Consumer Electronics, vol. 58, Issue. 2, May, 2012

Intra Prediction Algorithm Using Adaptive Modes

Kyungmin Lim, Jaeho Lee, Seongwan Kim, Daehyun Pak, Sangyoun Lee

The Journal of Korean Institute of Communications and Information Sciences, vol. 38A, Issue 6, page. 492-503, Jun, 2013

A Fast CU Size Decision Algorithm for HEVC

Jaeho Lee, Seongwan Kim, Kyungmin Lim, Sangyoun Lee

IEEE Transactions on Circuits and Systems for Video Technology, vol. 25, Issue. 3, July, 2014

Fast intermode decision algorithm based on general and local residual complexity in H.264/AVC

Jaeho Lee, Seongwan Kim, Kyungmin Lim, Sangyoun Lee

EURASIP Journal on Advances in Signal Processing, May, 2013

International Conferences

Fast CU Partitioning Skip Algorithm in HEVC Intra Prediction

Kyungmin Lim, Seongwan Kim, Jaeho Lee, Daehyun Pak, Changhyun Park, Sangyoun Lee

International Technical Conference on Circuits/Systems, Computers and Communications (ITC-CSCC), July, 2013

Efficient PU Skip Algorithm for Fast Intra Prediction

Kyungmin Lim, Sooyeon Kim, Jihun Oh, Sangyoun Lee

International Conference on Electronics, Information and Communication (ICEIC), February, 2013

Adaptive Crosstalk Reduction Method in Eye‐Tracking Stereoscopic Three‐Dimensional Displays Using Color Similarity and Inverse Filter

Young Min Kim, Kyungmin Lim, Jonghoon Lee, Hanjin Park, Youngjin Jo, Jaesung Lee

SID symposium digest of technical papers, August, 2025

An Approach to Reduce VR Sickness by Content Based Field of View Processing

Nupur Kala, Kyungmin Lim, Kwanghyun Won, Jaesung Lee, Sehoon Kim and Wonhee Choe

SID symposium digest of technical papers, June, 2017

Fast Stream Switch Method for Real-time Transfer Protocol

Daehyun Pak, Seongwan Kim, Kyungmin Lim, Changhyun Park and Sangyoun Lee

International Conference on Electronics, Information and Communication (ICEIC), January, 2015

Low-Delay Stream Switch Method for Real-Time Transfer Protocol

Daehyun Pak, Seongwan Kim, Kyungmin Lim, and Sangyoun Lee

The 18th IEEE International Symposium on Consumer Electronics (ISCE), June, 2014

Low Delay Video Streaming System for Interactive Broadcasting

Daehyun Pak, Seongwan Kim, Kyungmin Lim, and Sangyoun Lee

International Conference on Electronics, Information and Communication (ICEIC), January, 2014

Calibration between a Camera and Laser for 3D Modeling

Joongrock Kim, Sunjin Yu, Kyungmin Lim, Jinkyu Hwang, Sangheon Park and Sangyoun Lee

International Conference on Electronics, Information, and Communication (ICEIC), June, 2010

Thermal 3D Face Modeling System

Sunjin Yu, Jongrock Kim, Kyungmin Lim, Woowon Kim, and Sangyoun Lee

International Conference on Electronics, Information, and Communication (ICEIC), June, 2010

Fingertips Detection and Tracking based on Active Shape Models and an Ellipse

Sooyeon Kim, Younjung Park, Kyungmin Lim, Hyobin Lee, Sangki Kim, and Sangyoun Lee

TENCON, November, 2009

Domestic Conferences

Pedestrian Distance Estimation using a Single Camera Image

Taejae Jeon, Jinkyu Hwang, Kyungmin Lim, Eungyeol Song, Jaesung Kim, Kyungjae Lee, Sangyoun Lee

Summer Annual Conference of IEIE, June, 2014

Non-linear Rate-Distortion Optimization Model using the average of rate and distortion for HEVC

Changhyun Park, Seongwan Kim, Jaeho Lee, Kyungmin Lim, Daehyun Pak, Sangyoun Lee

Summer Annual Conference of IEIE, July, 2013

Low-latency architecture of x264 parallel coder

Daehyun Pak, Seongwan Kim, Jaeho Lee, Kyungmin Lim, Changhyun Park, Sangyoun Lee

신호처리합동학술대회, September, 2012

Face Recognition de-noising using wavelet transform and PCA for de-noising illumination

Changhyun Park, Seongwan Kim, Jaho Lee, Kyungmin Lim, Daehyun Pak, Sangyoun Lee

Summer Annual Conference of IEIE, June, 2012

Adaptive Syntax Addition Technique for Improving HEVC CAVLC Performance

Seongwan Kim, Jaeho Lee, Kyungmin Lim, Daehyun Pak, Sangyoun Lee

신호처리합동학술대회, September, 2011

cn switching method in run mode coding to improve HEVC CAVLC performance

Jaeho Lee, Seongwan Kim, Kyungmin Lim, Daehyun Pak, Sangyoun Lee

신호처리합동학술대회, September, 2011

Pose-invariant Face Detection using Color Information

Seongwan Kim, Jaeho Lee, Kyungmin Lim, Daehyun Pak, Sangyoun Lee

Summer Annual Conference of IEIE, July, 2009

Non-intrusive 3D Face Data Acquisition System

Joongrock Kim, Sunjin Yu, Kyungmin Lim, Sooyeon Kim, Sangyoun Lee

Summer Annual Conference of IEIE, June, 2008

US Patents

  • Electronic apparatus for displaying 3d image and operating method thereof (US20230326432A1)
  • Image processing device and method for operating same (US11995795B2)
  • Image processing device and operation method therefor (US11961203B2)
  • Electronic apparatus and control method thereof (US11575882B2)
  • Stacked display device and image providing method thereof (US12072567B2)

Contact

Please feel free to contact me for research collaboration or inquiries

✉
lkm1216@gmail.com