Hello! I am a Principal Researcher at Samsung AI Center (ex-SAIT AI 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 the interaction between humans and machine learning models, focusing on user-driven machine learning [J1,C3,C4], gestural user interfaces [C2], and the biomechanical analyses of gestural interactions [J2,C5]. More recently, my focus has shifted toward advancing multi-modal 3D scene understanding [C7-C10] , cross-modal representation learning [C7,C8], and addressing domain shift problems [C6,C8,C9] 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 skill learning.

Contact | Google Scholar | LinkedIn | CV

(∗: equal contributions, †: corresponding authors)

Conference Proceedings

• 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 (To appear, CVPR, 22.1% acceptance rate), 2025
• [paper]


• 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]


• 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] [bibtex]


• 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]


• 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]


• 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]


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


• 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]


• 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]


• 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]


• 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]

Journal Articles

• 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]


• 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]

Patents

∘ Granted

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


• Pressure Sensing Device Interface Representation
• US Patent US11320984B2, 2022


• Modifying an Image based on Identifying a Feature
• US Patent US11023769B2, 2021

∘ Published & Pending

• Method and Apparatus with Object Estimation Model Training
• US Patent App. 18/340,996, 2024


• Method and Apparatus with Object Detector Training
• US Patent App. 18/451,287, 2024


• Method and Apparatus with Data Labeling
• US Patent App. 18/109,928, 2023