Sujin's Homepage

Sujin Jang, Ph.D.

AI|Robotics Researcher

Hello! I am a Principal Researcher at Samsung AI Center (ex-SAIT AI Research Center) in Suwon, South Korea. Previously, I worked as a Machine Learning Researcher at Motorola Mobility LLC in Chicago, IL. Before that, I completed my Ph.D. at Purdue University under the guidance of Dr. Karthik Ramani, where I explored interactions among humans, robots, and machine learning models, focusing on user-driven machine learning [J1,C3,C4], gestural user interfaces [C2], and biomechanical analysis of gestural interactions [J2,C5]. More recently, my focus has shifted toward advancing multi-modal 3D scene understanding [C7-C11], cross-modal representation learning [C7,C8], and addressing domain shift problems [C6,C8,C9,C12,C13] in computer vision and robotic tasks. At a fundamental level, I am deeply interested in exploring the human-like adaptability and generalizability of physical AI agents capable of solving complex real-world problems, such as autonomous driving and robotic manipulation.

Contact | Google Scholar | LinkedIn | CV

Academic Publications

(C=Conference, J=Journal, P=Pre-print, ∗: Eequal contributions, †: Corresponding authors)

[P4] DAM-VLA: VLM-based Dynamic Action Model for Robot Manipulation
Xiongfeng Peng, Jiaqian Yu, Yamin Mao, Yi Zhou, Chao Zhang, Weiming Li, Sujin Jang, Dongwook Lee, Daehyun Ji

[P3] MoSE: Skill-by-Skill Mixture-of-Experts Learning for Embodied Autonomous Machines
Lu Xu, Jiaqian Yu, Xiongfeng Peng, Yiwei Chen, Weiming Li, Jaewook Yoo, Sunghyun Chung, Sujin Jang, Dongwook Lee, Daehyun Ji, Chao Zhang

[P2] Cross-Modal Domain Generalization for Multi-View 3D Object Detection
Gyusam Chang, Jiwon Lee, Sujin Jang, Jinkyu Kim, Dongwook Lee, Daehyun Ji, Wonjeong Ryoo, Sangpil Kim

[P1] Understanding Dimensional Collapse in Cross-Modal Feature Distillation
Dae Ung Jo^∗, Sujin Jang^∗, Jay Heo, Sung Ju Hwang

[C13] Active Test-time Vision-Language Navigation
Heeju Ko, Sungjune Kim, Gyeongrok Oh, Jeongyoon Yoon, Honglak Lee, Sujin Jang, Seungryong Kim, Sangpil Kim
Conference on Neural Information Processing Systems (NeurIPS, 24.5% acceptance rate), 2025
[arXiv]

[C12] Test-Time Adaptation for Online Vision-Language Navigation with Feedback-based Reinforcement Learning
Sungjune Kim^∗, Gyeongrok Oh^∗, Heeju Ko, Daehyun Ji, Dongwook Lee, Byung-Jun Lee, Sujin Jang^†, Sangpil Kim^†
International Conference on Machine Learning (ICML, 26.9% acceptance rate), 2025
[paper]

[C11] 3D Occupancy Prediction with Low-Resolution Queries via Prototype-aware View Transformation
Gyeongrok Oh^∗, Sungjune Kim^∗, Heeju Ko, Hyung-gun Chi, Jinkyu Kim, Dongwook Lee, Daehyun Ji, Sungjoon Choi, Sujin Jang^†, Sangpil Kim^†
Conference on Computer Vision and Pattern Recognition (CVPR, 22.1% acceptance rate), 2025
[paper] [project_page]

[C10] Unveiling the Hidden: Online Vectorized HD Map Construction with Clip-Level Token Interaction and Propagation
Nayeon Kim^∗, Hongje Seong^∗, Daehyun Ji, Sujin Jang^†
Conference on Neural Information Processing Systems (NeurIPS, 25.8% acceptance rate), 2024
[paper] [project_page] [bibtex]

[C9] Unified Domain Generalization and Adaptation for Multi-View 3D Object Detection
Gyusam Chang^∗, Jiwon Lee^∗, Donghyun Kim, Jinkyu Kim, Dongwook Lee, Daehyun Ji, Sujin Jang^†, Sangpil Kim^†
Conference on Neural Information Processing Systems (NeurIPS, 25.8% acceptance rate), 2024
[paper] [project_page] [code] [bibtex]

[C8] CMDA: Cross-Modal and Domain Adversarial Adaptation for LiDAR-based 3D Object Detection
Gyusam Chang^∗, Wonweok Roh^∗, Sujin Jang, Dongwook Lee, Daehyun Ji, Gyeongrok Oh, Jinsun Park, Jinkyu Kim, Sangpil Kim
AAAI Conference on Artificial Intelligence (AAAI, 23.7% acceptance rate), 2024
[paper] [arXiv] [supp] [video] [poster] [bibtex]

[C7] STXD: Structural and Temporal Cross-Modal Distillation for Multi-View 3D Object Detection
Sujin Jang^∗, Dae Ung Jo^∗, Sung Ju Hwang, Dongwook Lee, Daehyun Ji
Conference on Neural Information Processing Systems (NeurIPS, 26.1% acceptance rate), 2023
[paper] [slides] [poster] [bibtex]

[J2] Advanced Modeling Method for Quantifying Cumulative Subjective Fatigue in Mid-Air Interaction
Ana Villanueva^∗, Sujin Jang^∗, Wolfgang Stüerzlinger, Satyajit Ambike, Karthik Ramani
International Journal of Human-Computer Studies (IJHCS), Vol 169, 2023
[paper] [bibtex]

[C6] DaDA: Distortion-aware Domain Adaptation for Unsupervised Semantic Segmentation
Sujin Jang, Joohan Na, Dokwan Oh
Conference on Neural Information Processing Systems (NeurIPS-Oral, 184/2665~6.9%), 2022
[paper] [project_page] [slides] [poster] [bibtex]

[C5] Modeling Cumulative Arm Fatigue in Mid-Air Interaction based on Perceived Exertion and Kinetics of Arm Motion
Sujin Jang, Wolfgang Stüerzlinger, Satyajit Ambike, Karthik Ramanii
ACM Conference on Human Factors in Computing Systems (CHI, 25% acceptance rate), 2017
[paper] [slides] [code] [bibtex]

[J1] MotionFlow: Visual Abstraction and Aggregation of Sequential Patterns in Human Motion Tracking Data
Sujin Jang, Niklas Elmqvist, Karthik Ramani
IEEE Transaction on Visualization and Computer Graphics (TVCG), Vol 22. no.1, 2016, *Presented at IEEE Conference on Visual Analytics Science & Technology (VAST, 22% acceptance rate), 2015
[paper] [video-30sec] [video-full] [slides] [bibtex]

[C4] A Collaborative Filtering Approach to Real-Time Hand Pose Estimation
Chiho Choi, Ayan Sinha, Joon Hee Choi, Sujin Jang, Karthik Ramani
IEEE International Conference on Computer Vision (ICCV), 2015
[paper] [supp] [video] [bibtex]

[C3] GestureAnalyzer: Visual Analytics for Pattern Analysis of Mid-Air Hand Gestures
Sujin Jang, Niklas Elmqvist, Karthik Ramani
ACM Symposium on Spatial User Interaction (SUI, 29% acceptance rate), 2014
[paper] [video] [bibtex]

[C2] PuppetX: A Framework for Gestural Interactions With User Constructed Playthings
Saikat Gupta, Sujin Jang, Karthik Ramani
ACM Conference on Advanced Visual Interfaces (AVI, 28% acceptance rate), 2014
[paper] [video] [bibtex]

[C1] Experimental Results for Moving Object Structure Estimation using an Unknown Input Observer Approach
Sujin Jang, Ashwin Dani, Carl Crane, Warren Dixon
ASME Dynamic Systems and Control Conference (DSCC, Best Paper in Session Award), 2012
[paper] [video1] [video2] [bibtex]

Patents

Granted

Method and device with data processing using neural network, US12169917B2, 2024

Pressure Sensing Device Interface Representation, US11320984B2, 2022

Modifying an Image based on Identifying a Feature, US11023769B2, 2021

Published & Pending

Method and Apparatus with Object Detection Model Training, US20250157195A1, 2025

Method and Apparatus with Vector Map Learning and Generation, US20250086469A1, 2025

Method and Apparatus with Object Estimation Model Training, US20240211749A1, 2024

Method and Apparatus with Object Detector Training, US20240161442A1, 2024

Method and Apparatus with Data Labeling, US20230360381A1, 2023

Professional Experience

Samsung AI Center, Suwon, South Korea

Principal Researcher (Task Leader), Autonomous Machine Intelligence Team (Dec. 2024 - Present)

Samsung Advanced Institute of Technology (SAIT), Suwon, South Korea

Staff → Principal Researcher, Autonomous Driving Team (Jan. 2019 - Dec. 2024)

Motorola Mobility LLC., Chicago, IL, USA

Machine Learning Staff Researcher, Device+Machine Learning Team (Jun. 2017 - Dec. 2018)

Education

Purdue University, West Lafayette, Indiana, USA

Ph.D., Mechanical Engineering (Aug. 2017)

Specialization: Human-Computer Interaction; Machine Learning; Robotics

Advisor: Karthik Ramani

Thesis: Methods for Analyzing Natural Patterns and Physical Ergonomics of Human Gestures in Mid-Air Interaction

University of Florida, Gainesville, Florida, USA

M.S., Mechanical Engineering (Aug. 2012)

Specialization: Controls; Robotics; Machine Learning

Advisor: Carl D. Crane III and Warren E. Dixon

Thesis: Experimental Demonstration of Structure Estimation of Moving Objects Using Unknown Input Observers

Kookmin University, Seoul, South Korea

B.S., Mechanical and Automotive Engineering (Aug. 2010)

Undergraduate Research Assistant, Unmanned Vehicle Lab.

Advisor: Jungha Kim

Honors and Awards

Samsung Best Paper Award for "Distortion-aware Domain Adaptation" (AI/SW), Samsung Group, 2023

NeurIPS Oral Presentation (Top 6.9% of Submissions), Neural Information Processing Systems, 2022

Boundless Search for Breakthroughs Award for "Autonomous Driving", SAIT, Samsung Electronics, 2022

Device+ Team Excellence Award for "AR Vertical Healthcare", Lenovo, 2018

Estus H. and Vashti L. Magoon Award for Teaching Excellence, Purdue University, 2015

Awarded for excellent teaching in "ME444: Toy Design"

Best Paper in Session Award, ASME Dynamic Systems and Control Conference, 2012

Academic Service

[CVPR] Reviewer, IEEE Conference on Computer Vision and Pattern Recognition (2023,2024,2025)

[ICCV] Reviewer, CVF/IEEE International Conference on Computer Vision (2025)

[ECCV] Reviewer, European Conference on Computer Vision (2024)

[WACV] Reviewer, IEEE/CVF Winter Conference on Applications of Computer Vision (2026)

[NeurIPS] Reviewer, Conference on Neural Information Processing Systems (2025)

[AAAI] Reviewer, AAAI Conference on Artificial Intelligence (2025,2026)

[JCISE] Reviewer, ASME Journal of Computing and Information Science in Engineering (2023)

[CSCW] Reviewer, ACM Conference on Computer Supported Collaborative Work (2016,2018,2019,2024)

[CHI] Reviewer, ACM Conference on Human Factors in Computing Systems (2016,2018,2019,2021,2024)

[ISMAR] Reviewer, IEEE International Symposium on Mixed and Augmented Reality (2021)

[VR] Reviewer, IEEE Conference on Virtual Reality and 3D User Interfaces (2018,2019)

[3DUI] Reviewer, IEEESymposium on 3D User Interfaces (2016)

[TVCG] Reviewer, IEEE Transaction on Visualization and Computer Graphics (2017)

[VAST] Reviewer, IEEE Conference on Visual Analytics Science and Technology (2016,2017,2018,2019,2020)

[InfoVis] Reviewer, IEEE Conference on Information Visualization (2015)

[EuroVis] Reviewer, EG/VGTC Conference on Data Visualization (2016)

[VIS] Student Volunteer (2015)

[UIST] Reviewer, ACM Symposium on User Interface Software and Technology (2018)

[MobileCHI] Reviewer, ACM International Conference on Mobile Human-Computer Interaction (2019)

[SUI] Reviewer, ACM Symposium on Spatial User Interaction (2019)

[DIS] Reviewer, ACM Conference on Designing Interactive Systems (2016,2017)

[TEI] Reviewer, ACM Conference on Tangible, Embedded, and Embodied Interaction (2017)

[IDC] Reviewer, ACM Conference on Interaction Design and Children (2017)

[VRST] Reviewer, ACM Conference on Virtual Reality Software and Technology (2017)

Last updated: