Dongsik Yoon

Curriculum Vitae Last updated October 2026
kevinds1106@korea.ac.kr
https://stillrot.github.io/
Research
Interest
Real-World Computer Vision (Action Recognition, Anomaly Detection in CCTV)
Synthetic Data Generation via Generative Models
Edge AI (Resource-Efficient Inference, On-Device Deployment)
Education
M.S., Electrical Engineering
Korea University, Seoul, Korea
Dissertation: Deep Learning Framework Design via Condition Guided Learning
Advisor: Prof. Hanseok Ko (Intelligent Signal Processing Laboratory, ISPL)
Mar. 2021 - Feb. 2023
B.S., ICT, Multimedia Engineering
Baekseok University, Cheonan, Korea
Advisor: Prof. No-Yoon Kwak (Image Processing and Computer Graphics Laboratory, IPCG)
Overall GPA 4.36 / 4.5, Major GPA 4.49 / 4.5
Mar. 2015 - Feb. 2021
Industry
Experience
AI Research Engineer
HDC LABS, Seoul, Korea
Research on CCTV-based vision systems for indoor safety (fire detection, elevator analytics)
May 2024 - Present
AI Research Engineer
Metaverse Entertainment, Seoul, Korea
Generative AI for virtual-influencer content (image editing, voice synthesis)
Jan. 2023 - May 2024
Publications
International Conference
D. Yoon, "Ask Before It Tells: Benchmark-to-Robot Body-Cue Transfer for a Question-First Bedside Robot", IEEE/RSJ International Conference on Intelligent Robots and Systems Workshop (IROSW), 2026
H. Park, J. Kim, J. Kim, D. Yoon, "CHOP: Chunkwise Context-Preserving Framework for RAG on Multi Documents", arXiv preprint, 2026
D. Yoon, J. Kim, B. Park, S. Kim, S. Hwang, "Real-World Videos Are Not Silent: Synthesized Audio Dataset Augmentation for Multimodal Action Recognition", Under review, 2026
D. Yoon*, J. Kim*, D. Lee*, "Resource-Efficient RGB-Only Action Recognition for Edge Deployment", arXiv preprint, 2026
D. Yoon, J. Kim, "From Prompts to Deployment: Auto-Curated Domain-Specific Dataset Generation via Diffusion Models", IEEE/CVF Winter Conference on Applications of Computer Vision Workshop (WACVW), 2026
D. Yoon, J. Kim, "Your Super Resolution Model is not Enough for Tackling Real-World Scenarios", International Conference on Computer Vision Workshop (ICCVW), 2025
J. Kim*, Y. Lee*, D. Yoon*, C. Jung, G. Lee, "An Integrated YOLO and VLM System for Fire Detection in Enclosed Environments", International Conference on Learning Representations Workshop (ICLRW), 2025
D. Yoon, J. Kim, S. Song, Y. Lee, G. Lee, "OverallNet: Scale-Arbitrary Lightweight SR Model for handling 360° Panoramic Images", SIGGRAPH Asia Posters, 2024
J. Kwak, Y. Li, D. Yoon, D. Kim, H. Ko, "Injecting 3D Perception of Controllable NeRF-GAN into StyleGAN for Editable Portrait Image Synthesis", European Conference on Computer Vision (ECCV), 2022
D. Yoon, J. Kwak, Y. Li, D. Han, H. Ko, "DIFAI: Diverse Facial Inpainting using StyleGAN Inversion", IEEE International Conference on Image Processing (ICIP), 2022
J. Kwak, Y. Li, D. Yoon, D. Han, H. Ko, "Generate and Edit Your Own Character in a Canonical View", IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshop (CVPRW), AI for Content Creation Workshop, 2022
D. Yoon, J. Kwak, Y. Li, Y. Jin, D. Han, H. Ko, "Reference Guided Image Inpainting using Facial Attributes", British Machine Vision Conference (BMVC), 2021
J. Kwak, Y. Jin, Y. Li, D. Yoon, D. Kim, H. Ko, "Adverse Weather Image Translation with Asymmetric and Uncertainty-aware GAN", British Machine Vision Conference (BMVC), 2021
Y. Li, Y. Jin, J. Kwak, D. Yoon, D. Han, H. Ko, "Adaptive Content Feature Enhancement GAN for Multimodal Selfie to Anime Translation", British Machine Vision Conference (BMVC), 2021
International Journal
Y. Li, D. Yoon, B. Ku, H. Ko, "ConSeisGen: Controllable Synthetic Seismic Waveform Generation", IEEE Geoscience and Remote Sensing Letters (GRSL), 2023 (IF 4.8, SCI-E)
D. Yoon, J. Kim, V. Lorant, S. Kang, "Manipulation of Age Variation Using StyleGAN Inversion and Fine-Tuning", IEEE Access, 2023 (IF 3.9, SCI-E)
D. Yoon, Y. Li, B. Ku, H. Ko, "Estimation of Magnitude and Epicentral Distance from Seismic Waves using Deeper CRNN", IEEE Geoscience and Remote Sensing Letters (GRSL), 2023 (IF 5.3, SCI-E)
Projects
In-elevator situational awareness
Jan. 2025 - Feb. 2026
Built a PoC for in-cab anomaly detection (pet detection, occupancy estimation) that triggered a HDC LABS / HDC Hyundai Industrial Development / Hyundai Elevator three-party MOU. Constructed a domain-specific synthetic + real labelling pipeline and a deployment-automation environment.
Affiliation: HDC LABS
AI fire / smoke detection for underground parking lots
May 2024 - Oct. 2025
Designed a synthetic-data pipeline for CCTV-specific fire scenes; trained YOLOv8s + Florence2 VLM detector and migrated to a YOLOv8n + CLIP edge variant for Hailo dual-chip deployment. Demonstrated at Gwangmyeong Central IPARK.
Affiliation: HDC LABS
Designer-facing image generation for virtual influencers
Jan. 2023 - May 2024
Curated Stable Diffusion LoRA workflows and a Gradio-based internal UI used for MAVE: / rina.8k content. Researched a two-stage StyleGAN-based face re-aging framework, published in IEEE Access.
Affiliation: Metaverse Entertainment
Expressive multi-speaker TTS
Jan. 2022 - Dec. 2022
Built reference-audio-conditioned emotion control model for reading-style / dialog-style TTS (MOS +5%, Similarity-MOS +11% over baseline, n=30).
Funding: Deep Machine Laboratory (DMLAB)
Generative virtual earthquake scenarios for the Korean Peninsula
Jan. 2021 - Dec. 2021
Built a GAN-based generator and verification pipeline producing M ≥ 3.0 seismic waveforms; delivered a scenario-authoring software to KMA.
Funding: Korea Meteorological Administration (KMA)
Patents
13 KR patents granted, 4 patent applications in progress (3 KR + 1 US)
2024 - 2026
Topics include fire detection in enclosed environments, in-elevator pet / occupancy analytics, construction-site hazard detection, domain-specific synthetic data generation, personalized object tracking, and AI-driven home agents.
See stillrot.github.io/#patents for the full list.
Academic
Service
Reviewer (peer-review)
ICIP 2026; ICBINB @ ICLR 2026; TAG @ CoRL 2026
Instructor, K-Digital Training | HDC LABS
Generative AI for Analytics and Service Development
2025 - 2026
Grants & Awards
Best Paper Presentation Award
2020
Korea Multimedia Society (KMS) Summer Annual Conference
Best Paper Award
2019
Joint Conference of Korea Digital Contents Society (KDCS) and Korea Institute of Information Technology (KIIT)
Gold Prize, Undergraduate Paper Competition
2019
KIIT / KDCS Joint Summer Conference
Encouragement Award
2019
Korean Institute of Communications and Information Sciences (KICS) Winter Annual Conference