Publications

2024

• TIE-KD: Teacher-Independent and Explainable Knowledge Distillation for Monocular Depth Estimation [preprint][code]

  • Sangwon Choi, Deajune Choi, Duksu Kim

  • arXiv, 2024.14340 (under review for a journal)

• HoloSR: End-to-End 3D Hologram Super-Resolution with Deep Neural Network [preprint]

  • You chan No, Jaehong Lee, Han-Ju Yeom, Sung Min Kwon, Duksu Kim

  • Optica Open (under review for a journal)

• 다중 UGV를 활용한 임무 영역 탐사를 위한 공간 분할과 경로 계획 (Space Partitioning and Path Planning for Mission Area Exploration using Multiple UGVs) (SCOPUS) [paper][video]

  • 최상원, 이상현, 우상범, 김태형, 설형수, 김덕수

  • Journal of Institute of Control, Robotics and Systems, Vol 30, No. 3, pp. 197-205, 2024

2023

• Out-of-core diffraction algorithm using multiple SSDs for ultra-high-resolution hologram generation (SCI(Q1), IF: 3.8) [paper][bibtex]

  • Jaehong Lee, Duksu Kim

  • Optics Express, Vol 31, No. 18, pp. 28683-28700, 2023

• Large-scale 3D-FFT computation on GPU (SCI, IF: 1.622) [paper][bibtex]

  • Jaehong Lee, Duksu Kim

  • ETRI Journal, Vol. 24, No. 6, pp. 1035-1045, 2023

• 세그멘테이션 기반 차선 인식 네트워크를 위한 적응형 키포인트 추출 알고리즘 (Adaptive Key-point Extraction Algorithm for Segmentation-based Lane Detection Network) (KCI) [paper]

  • 이상현, 김덕수

  • Journal of the Korea Computer Graphics Society, Vol.  29, No. 1, pp. 1-11, 2023

• 회절연산 정밀도에 따른 CGH 기반 홀로그램 생성 품질 분석 (Quality Analysis on Computer Generated Hologram Depending on the Precision on Diffraction Computation) (KCI) [paper]

  • 이재홍, 김덕수

  • 방송공학회논문지, Vol 28, No. 1, pp. 21-30, 2023

2022

• Real-time lens distortion algorithm on an edge device with GPU (SCIE(Q1), IF: 3.367) [paper][bibtex]

  • Young-Woo Kim, Hyeon-seok Yang, Duksu Kim

  • IEEE Access, Vol. 10, pp. 41748-41757, 2022

• Real-time lens distortion algorithm on embedded GPU systems (Poster) [extended abstract]

  • Young-Woo Kim, Duksu Kim

  • Poster session, SIGGRAPH 2022

2021

• HI-FFT: Heterogeneous Parallel In-place Algorithm for Large-scale 2D-FFT (SCIE(Q1), IF: 3.367) [paper][bibtex]

  • Homin Kang, Jaehong Lee, Duksu Kim

  • IEEE Access, Vol. 9, pp. 120261-120273, 2021

• Out-of-Core GPU 2D-Shift-FFT algorithm for Ultra-High-Resolution Hologram Generation (SCIE(Q1), IF: 3.894) [paper][bibtex]

  • Jaehong Lee, Homin Kang, Han-Ju Yeom, Sanghoon Cheon, Joongki Park, Duksu Kim

  • Optics Express, Vol. 29, No. 12, pp. 19094-19112, 2021

• RGB-D 이미지 기반 물체 자세 인식 딥 러닝 파이프라인 (Deep-Learning Pipeline for Object Pose Estimation from an RGB-D Image) (SCOPUS) [paper (Koraen)]

  • 노유찬, 김영우, 김대건, 한현규, 송영기, 김덕수

  • Journal of Institute of Control, Robotics and Systems, Vol. 27. No. 8, pp. 593-601, 2021

• 자율 주행 자판기 로봇 설계 및 구현 (Design and Implementation of Self-Driving Vending Robot) (SCOPUS) [paper (Korean)]

  • Daegun Kim, Sang Won Choi, Duksu Kim

  • Journal of Institute of Control, Robotics and Systems, Vol. 27, No. 6, pp. 411-417, 2021

• High-performance 2D-FFT Computation for Ultra-high-resolution Hologram Generation

  • Jaehong Lee, Homin Kang, Han-Ju Yeom, Sanghoon Cheon, Joongki Park, Duksu Kim

  • Invited paper, 3D Systems and Application (3DSA), 2021

  • Received the best paper award

• 3차원 형상 복원을 위한 점진적 점유 예측 네트워크 (Progressive occupancy network for 3D reconstruction) (KCI) [paper (Korean)][code]

  • 김용규, 김덕수

  • Journal of the Korea Computer Graphics Society, Vol. 27, No. 3, pp. 65-74, 2021

• Development of Recommendation Model for Image Keypoint Detection and Descriptor Extraction Algorithm (KCI) [paper (Korean)]

  • 김영우, 김덕수, 김승희

  • Journal of KIIT. Vol. 19, No. 4, pp. 27-35, 2021

• Suggestion of RE and TDD-based V&V Development Process for Scientific Software Implementation (KCI) [paper (Korean)]

  • 이재홍, 김덕수, 김승희

  • The Journal of The Institute of Internet, Broadcasting and Communication, Vol. 21, No. 2, pp. 79-88, 2021

2020

• • HPMaX: Heterogeneous Parallel Matrix Multiplication using CPUs and GPUs (SCIE, IF: 2.220) [paper][author preprint][bibtex]

    • Homin Kang, Hyuck Chan Kwon, Duksu Kim

    • Computing, Vol. 102, No. 12, pp. 2607-2631, Dec., 2020

  • Accurate and Efficient GPU Ray-Casting Algorithm for Volume Rendering Unstructured Grid Data (SCIE, IF: 1.347) [Paper][bibtex]

    • Gibeom Gu, Duksu Kim

    • ETRI Journal, Vol. 42, No. 4, pp. 608-618, Aug., 2020

  • Parallel Cell-Connectivity Information Extraction Algorithm for Ray-casting on Unstructured Grid Data (KCI) [paper (Korean)] [bibtex]

    • 이지훈, 김덕수 (Duksu Kim)

    • Journal of the Korea Computer Graphics Society, Vol. 26, No.1, pp. 17-25, 2020 

  • CNN Architecture Predicting Movie Rating from Audience’s Reviews Written in Korean (KCI) [paper] [bibtex]

    • 김형찬, 오흥선, 김덕수 (Duksu Kim)

    • KIPS Trans. on Computer and Communication Systems, Vol. 9, No. 1, pp. 17-24, 2020

      • Selected as one of the highlight papers for the first half of 2020 [link]

  •  GPU-based Shift-FFT Implementation for Ultra-High Resolution Hologram Generation

    • 이재홍, 강호민, 전상훈, 염한주, 박중기, 김덕수

    • 한국방송미디어공학회 하계 학술대회, 2020

2019

• • Visualization Pipeline Design and Implementation for Cluster-based Scientific Visualization Tools [paper (Korean)] (KCI)

    • (클러스터 기반 과학적 가시화 도구를 위한 가시화 파이프라인 설계 및 구현)

    • 이중연, 김덕수 (Duksu Kim)

    • KIISE Transactions on Computing Practices, Vol. 25, No. 6, pp. 285-293, 2019

  • Memory Efficient Parallel Ray Casting Algorithm for Unstructured Grid Volume Rendering on Multi-core CPUs [paper]

    • Gibeom Gu, Duksu Kim

    • EuroVis (Poster) 2019

2009~ 2017 (Before Prof. Kim joined at KOREATECH) 

• • Out-of-Core Proximity Computation for Particle-based Fluid Simulations [project page]

    • Duksu Kim, Myung-Bae Son, Young J. Kim, Jeong-Mo Hong, Sung-Eui Yoon 

    • High Performance Graphics, 2014 

    • Invited to submit extended version to computer graphics forum

  • Scheduling in Heterogeneous Computing Environments for Proximity Queries [project page] (SCI, IF: 3.780)

    • Duksu Kim, Jinkyu Lee, Junghwan Lee, InSik Shin, John Kim, Sung-eui Yoon

    • IEEE Transactions on Visualization and Computer Graphics (TVCG), Setp., 2013 

    • the Spotlight Paper for the September issue 

  • Multi-Resolution Cloth Simulation [project page] (SCIE, IF: 2.116)

    • YongJoon Lee, Sung-eui Yoon, SeungWoo Oh, DukSu Kim, SungHee Choi 

    • Computer Graphics Forum+ (Pacific Graphics), 2010 

  • FASTCD: Fracturing-Aware Stable Collision Detection [project page]

    • Jae-Pil Heo, Joon-Kyung Seong, DukSu Kim, Miguel A. Otaduy, Jeong-Mo Hong, Min Tang, Sung-Eui Yoon 

    • ACM SIGGRAPH/Eurographics Symp. on Computer Animation (SCA), 2010 

  • RACBVHs: Random-Accessible Compressed Bounding Volume Hierarchies [project page] (SCI, IF: 3.780)

    • Tae-Joon Kim, Bochang Moon, Duksu Kim, and Sung-Eui Yoon

    • IEEE Transactions on Visualization and Computer Graphics (TVCG), 2010 

  • HPCCD: Hybrid Parallel Continuous Collision Detection using CPUs and GPUs  [project page] (SCIE, IF: 2.116)

    • DukSu Kim, Jae-Pil Heo, JaeHyuk Huh, John Kim, and Sung-Eui Yoon

    • Computer Graphics Forum+ (Pacific Graphics) 2009 

    • Received a distinguished paper award at the conf.)

  • Memory Efficient Parallel Ray Casting Algorithm for Unstructured Grid Volume Rendering on Multi-core CPUs  [paper (Korean)] (KCI 등재)

    • (비정렬 격자 볼륨 렌더링을 위한 다중코어 CPU기반 메모리 효율적 광선 투사 병렬 알고리즘 ) 

    • 김덕수 (Duksu Kim) 

    • Journal of KIISE, Vol. 43, No. 3, 2016 

  • Artificial Potential Function for Driving a Road with Traffic Light [paper (Korean)] (KCI 등재)

    • (신호등 신호에 따른 차량 주행 제어를 위한 인공 전위 함수) 

    • 김덕수(Duksu Kim) 

    • Journal of KIISE, Vol. 42, No. 10, 2015 

  • Urban Mobility Simulation [paper (Korean)] (KCI 등재)

    • (도심 교통 시뮬레이션) 

    • 김경아, 김덕수(Duksu Kim), 윤성의 

    • 한국컴퓨터그래픽스학회/컴퓨터그래픽스 논문지, Vol. 17, No. 4, 2011 

Ph. D. dissertation

• 그래픽스 응용프로그램을 위한 이종 연산장치 기반 근접질의 병렬처리 기술 (Parallel Proximity Computation on Heterogeneous Computing Systems for Graphics Applications)

  • 김덕수 (Duksu Kim), 전산학부, KAIST, 2014

  • Talk slide

MS Thesis

• • 경량 단안 깊이 추정 네트워크 성능 향상을 위한 설명 가능한 특징 맵 기반 선생 모델 독립적 지식 증류 방법 (Explainable feature-map-based teacher-independent knowledge distillation framework for a lightweight monocular depth estimation network)

    • 최상원, 컴퓨터공학부, 한국기술교육대학교, 2024. 02

  • 홀로그램 업스케일링 딥러닝 네트워크 (Hologram super-resolution deep learning network)

    • 노유찬, 컴퓨터공학부, 한국기술교육대학교, 2023. 02

  • 컴퓨터 생성 홀로그래피를 위한  성능 대규모 고속 푸리에 변환 (High-performance large scale FFT for computer generated holography)

    • 이재홍, 컴퓨터공학부, 한국기술교육대학교, 2022. 02

  • GPU기반 엣지 디바이스를 이용한 실시간 렌즈 왜곡 보정 알고리즘 (Real-time lens distortion algorithm on an edge device with GPU)

    • 김영우, 컴퓨터공학부, 한국기술교육대학교, 2022. 02

  • 행렬 연산을 위한 고성능 이종 병렬처리 알고리즘강호민 (High-performance Heterogeneous parallel algorithms for Matrix operations)

    • 강호민, 컴퓨터공학부, 한국기술교육대학교, 2021. 08

  • 3차원 형상 복원을 위한 점진적 점유 예측 네트워크 (Progressive occupancy network for 3D reconstruction)

    • 김용규, 컴퓨터공학부, 한국기술교육대학교, 2021.08

    • Code