Sung-Un Park
박 성언 (朴 成彦)
park@stoz.kr
Staff Researcher, Samsung Advanced Institute of Technology
[Linkedin] [Google Scholar] [Naver]
In memoria aeterna erit iustus. ab auditione mala non timebit.
TECHNICAL AREAS
On-Device Artificial Intelligence, Computer Vision, Machine Learning, Biometrics, Industrial AI, Machine Vision
EDUCATION
Korea Advanced Institute of Science and Technology, Daejeon, Republic of Korea, Aug 2021-Present
Ph.D. Student, in Electrical Engineering.
Advisor: Prof. Changick Kim
Computational Intelligence Lab.
Kyung Hee University, Yongin, Republic of Korea, Mar 2015-Feb 2017
M.S., in Biomedical Engineering.
Thesis Title: A Depth Camera-based Human Activity Recognition via Deep Learning Recurrent Neural Network
Advisor: Prof. Tae-Seong Kim
Bio-imaging & Brain Engineering Lab.
Kyung Hee University, Yongin, Republic of Korea, Mar 2009-Feb 2015
B.S., in Biomedical Engineering & B.S., in Computer Engineering.
Served and honorably discharged upon completing military service from the 8th Mechanized Infantry Division, the Republic of Korea Army as a sergeant. (Oct 2010-Jul 2012)
Student Council President - School of Liberal Arts, College of Liberal Arts, Kyung Hee University. (2009)
Samsung Software Membership (SSM 23th,2013-2015)
PROFESSIONAL EXPERIENCE
Samsung Advanced Institute of Technology (SAIT), Samsung Electronics, Suwon, Republic of Korea, Feb 2017-Present
Staff Researcher, Machine Learning Tech. Unit, AI Research Center.
PROJECTS
Machine Vision, Jan 2022-Present
3D Scanning Electron Microscope Metrology [1]
Biometrics Anti-Spoofing System for Galaxy S series, Sep 2019-Sep 2022
Designed the software architecture of the Anti-Spoofing System.
Developed and implemented an algorithm to improve anti-spoofing performance for various user environments. (United States Patent Application 17/172,429)
Bixby Wake-up and Unlock System (Speaker Recognition) for Galaxy series, Jul 2018-Aug 2019 [1]
Designed the software architecture of the Speaker Recognition System.
Developed and implemented an algorithm to improve speaker recognition performance for various noise environments. (United States Patent Application 16/781,176)
Developed and implemented a user enrollment data update algorithm for robust speaker recognition. (United States Patent Application 16/872,548)
Intelligent Scan (Combines Iris Scan and Face Recognition) for Galaxy S9 and Note 9, Aug 2017-June 2018 [1] [2]
Participated in software architecture design of the Intelligent Scan System. (United States Patent Application 16/269,990)
Developed and implemented a multi-modal Face liveness detection algorithm to improve the anti-spoofing system. (United States Patent 11,023,757)
Developed and implemented a landmark alignment algorithm for weakly synchronized images with different characteristics as like field of view, resolution, spectrum to reduce time for landmark detection. (United States Patent Application 16/148,129)
Developed and implemented an algorithm to reduce face recognition time. (United States Patent 10,853,678)
Implemented an image filter and pre-processing code for low-light images to enhance performance at low-light environment. (United States Patent Application 16/223,382, European Patent EP3525165)
Researched hard sample mining algorithm for improving the performance of deep learning.
Face Recognition for Galaxy S8 and Note 8, Mar 2017-Jul 2017
Participated in face recognition system development.
Integrated code into a mobile system to release.
TECHNICAL EXPERIENCE
Bio-Imaging Lab., Kyung Hee University, Yongin, Republic of Korea, May 2013-Jan 2017
Depth Imaging Sensor based Human Activity Recognition
Developed a system that recognizes human activities (as like walking, jumping, raising arms, etc.). Used Recurrent Neural Network with Tensorflow to recognize human activities with joint angle information from depth image sequences.
INDUSTRIAL PROJECTS
Automatic TFT-LCD panel defect detection and classification system (with Samsung Display), Jul 2016-Dec 2016
Developed an automatic system that finds defect in the production process of the TFT-LCD panel and classifies them using Convolutional Neural Network.
Real-time detection and classification implemented.
Human Image Retrieval System via Convolutional Neural Network (with Mappers corporation), Nov 2015-Mar 2016
Developed an image retrieval system using Convolutional Neural Network to classify images containing humans in Tensorflow.
INTERNSHIP EXPERIENCE
Samsung Software Membership, Samsung Electronics, Suwon, Republic of Korea, Jan 2013-Mar 2015
Software Engineer
Developed trading business information applications to give technical barriers to trade (TBT) information and to find harmonized system (HS) codes. Awarded for excellent development in 2015 Technical Standard Idea Software Contest, April 2015, the Korean Agency for Technology and Standards, Republic of Korea (2015).
Developed an online sticky note application to make it easier to brainstorm. It is an electronic meeting room where multiple users can view and edit sticky notes synchronized (2014, funded by Visual Display Business, Samsung Electronics).
Developed an ARS speech-to-text application to visualize ARS voice. Awarded for Samsung Software Membership Tech Fair 2014 Excellent Project (2014). [youtube]
Developed an Android MAME emulator to can play with others by WIFI direct connection (2013).
HONORS AND AWARDS
Samsung Best Paper Award, Silver Award, "Fast and Robust Fingerprint Anti-Spoofing for Mobile Devices via Joint Contrastive Learning", Samsung Electronics, 2021. 10.
Samsung Best Paper Award, Bronze Award, "Bayesian Speaker Verification using Orthogonal Deep Representation", Samsung Electronics, 2019. 10.
2019년 1분기 DS부문 Future Creator상, Samsung Electronics, 2019.
The Future Creator Award (Device Solution) in the first quarter of 2019.
2017년 하반기 무한탐구상(혁신부문), Samsung Advanced Institute of Technology, 2018.1.
The Boundless search for breakthroughs Award (Innovation) in the second half of 2017.
2015 기술표준 아이디어 공모전 개발부분 우수상, 국가기술표준원, 2015
Award for Excellent Development in '2015 Technical Standard Idea Software Contest, April 2015, the Korean Agency for Technology and Standards, Republic of Korea.
Samsung Software Membership Technology Fair 2014 - Excellent Project
Our team was selected as regional representatives and participated in the '2014 Samsung Software Membership Technology Fair.
Award for Excellent Project '2014 Samsung Software Membership Technology Fair, Sep 2014, Samsung Electronics.
PUBLICATIONS
S. U. Park, J. H. Park, M. A. Al-masni, M. A. Al-antari, Md. Z. Uddin and T. -S. Kim, “A Depth Camera-based Human Activity Recognition via Deep Learning Recurrent Neural Network for Health and Social Care Services” in Proceedings of the Conference on Health and Social Care Information Systems and Technologies, HCist 2016, October 5-7, 2016. [link]
M. A. Al-antari, M. A. Al-masni, S. U. Park, J. H. Park, M. K. Metwally, Y. M. Kadah, S. M. Han and T. -S. Kim, “An Automatic Computer-Aided Diagnosis System for Breast Cancer in Digital Mammograms via Deep Belief Network,” Journal of Medical and Biological Engineering, vol. 38, 2018, pp. 443-456. [link]
S. B. Nam, S. U. Park, J. H. Park, M. Z. Uddin and T. -S. Kim, “Accurate 3D Human Pose Recognition via Fusion of Depth and Motion Sensors,” International Journal of Future Computer and Communication, vol. 4, 2015, pp. 336.
M. A. Al-masni, S. U. Park, J. H. Park, Y. M. Kadah, S. M. Han, T. -S. Kim and M. A. Al- antari, “Automatic Computer-Aided Diagnosis of Breast Cancer in Digital Mammograms via Deep Belief Network" in Proceedings of the Global Conference on Engineering and Applied Science, 19-21 July 2016, Hokkaido, Japan, 2016, pp. 1306-1314.
J. H. Park, S. U. Park, M. Z. Uddin, M. A. Al-antari, M. A. Al-masni and T. -S. Kim, “A Single Depth Sensor Based Human Activity Recognition via Convolutional Neural Network,” in Proceedings of the 6th International Conference on the Development of Biomedical Engineering, 27-29 June 2016, Ho Chi Minh City, Vietnam, 2016, pp. 329-332.
S. B. Nam, S. U. Park, J. H. Park and T. -S. Kim, “A Single Depth Sensor Based Human Activity Recognition via Deep Belief Network,” in Proceedings of the 4th World Conference on Applied Sciences, Engineering and Technology 24-26 October 2015, Kumamoto University, Japan, 2015, pp. 15-19.
H. J. Jeon, S. B. Nam, S. U. Park, J. H. Park, C. Yoo, J. T. Kim and T. Kim, “Unsupervised 3D human pose recognition from a single depth human silhouette using a geodesic map and kinematic body model," in Proceedings of the 9th International Conference on Ubiquitous Information Management and Communication, 2015, pp. 70.
S. B. Nam, S. U. Park, H. S. Han, H. J. Jeon, S. M. Kim and T. -S. Kim, “Improved 3D human skeletal pose reconstruction using the probability distribution of random forests in body parts recognition and joint proposals estimation,” in The Korea Society of Medical & Biological Engineering 2013 Nov 48, pp. 724-725.
M. A. Al-antari, M. A. Al-masni, S. U. Park, J. H. Park, Y. M. Kadah, S. M. Han and T. -S. Kim, “A novel computer-aided diagnosis system for breast cancer in digital mammograms via deep belief network,” in The Korea Society of Medical & Biological Engineering 2016 May 51, pp. 397-398.
B. Wei, M. Yang, T. Zhang, X. Tang, X. Huang, K. Kim, J. Lee, K. Cho and S. U. Park, "End-to-End Transformer-Based Open-Vocabulary Keyword Spotting with Location-Guided Local Attention," in The Proc. Interspeech 2021, 30 Aug-3 Sep 2021, Brno, Czechia, 2021, pp. 361-365.
PATENTS
S. Han, “Object recognition method and apparatus,” United States Patent 10,853,678 B2, 2020. [link, KR;US;CN;EP;JP]
S. Park, “Method and apparatus with liveness verification,” United States Patent 11,023,757 B2, 2021. [link]
S. Park, “Device and method with image matching,” United States Patent Application 16/148,129, 2019. [link, KR;US;CN;EP]
S. Park, “Method and apparatus with speaker recognition registration,” United States Patent Application 16/781,176, 2020. [link, US;CN;EP]
S. Park, “Method and device with data recognition,” United States Patent Application 16/845,464, 2020. [link, KR;US]
M. Ko, “Method and apparatus with image fusion,” United States Patent Application 16/223,382, 2019. [link, KR;CN;US;EP]
J. Kim, “Method and apparatus with selective combined authentication,” United States Patent Application 16/269,990, 2019. [link, US;EP;CN]
S. Park, “Method and apparatus with registration for speaker recognition,” United States Patent Application 16/872,548, 2020. [link, KR;US]
S. Park, “Method and apparatus with anti-spoofing,” United States Patent Application 17/172,429, 2020. [link, US;CN;EP;JP]
S. Cho, “Apparatus with fake fingerprint detection and method with updating of model to prevent fake fingerprint,” United States Patent Application 17/125,273, 2020.
J. Kim, “Method and apparatus with fake fingerprint detection,” United States Patent Application 17/137,163, 2020.
S. Cho, “Method and apparatus of detecting whether biometric information is spoofed,” Korea Patent Application 10-2020-0172382, 2020.
Extracurricular