I am a Senior Engineer at Qualcomm AI Research in Korea. I received my Ph.D in Artificial Intelligence from Korea Advanced Institute of Science and Technology (KAIST) under the supervision of Professor Jaegul Choo. My research interests include image & video generation, image-to-image translation, and data scarcity problems. In particular, I am interested in developing deep learning models for virtual try-on, head reenactment, hairstyle transfer, and long-tailed learning. Recently, I focus on developing algorithms to personalize the foundation models for real-world applications.
[Feb. 2024] StableVITON paper accepted to CVPR 2024. [Dec. 2023] I joined Qualcomm AI Research as a senior engineer. [Dec. 2023] Two papers accepted to ICASSP 2024. [Dec. 2023] New Normal paper accepted to AAAI 2024. [Nov. 2023] I successfully defended my PhD thesis! [Jul. 2023] Label Shift Adapter paper accepted to ICCV 2023. [May. 2023] Paint-by-Sketch paper accepted to AI4CC CVPRW 2023.
Balanced Learning for Multi-Domain Long-tailed Speaker Recognition
Janghoon Cho*, Sunghyun Park*, Hyunsin Park, Hyoungwoo Park, Seunghan Yang, and Sungrack Yun
IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2024, Seoul, Accepted
[Paper]
When Model Meets New Normals: Test-time Adaptation for Unsupervised Time-series Anomaly Detection
Dongmin Kim, Sunghyun Park, and Jaegul Choo
AAAI Conference on Artificial Intelligence (AAAI), 2024, Vancouver, Accepted.
[Paper]
Label Shift Adapter for Test-Time Adaptation under Covariate and Label Shifts
Sunghyun Park, Seunghan Yang, Jaegul Choo, and Sungrack Yun
International Conference on Computer Vision (ICCV), 2023, Paris, Accepted.
[Paper]
AnimeCeleb: Large-Scale Animation CelebHeads Dataset for Head Reenactment
Kangyeol Kim*, Sunghyun Park*, Jaeseong Lee*, Sunghyo Chung, Junsoo Lee, and Jaegul Choo
European Conference on Computer Vision (ECCV), 2022, Tel Aviv, Accepted
Best Paper Award, Korean Artificial Intelligence Association 2022.
[Paper]
[Code & Dataset]
Style Your Hair: Latent Optimization for Pose-Invariant Hairstyle Transfer via Local-Style-Aware Hair Alignment
Chaeyeon Chung*, Taewoo Kim*, Yoonseo Kim*, Sunghyun Park, Kangyeol Kim, and Jaegul Choo
European Conference on Computer Vision (ECCV), 2022, Tel Aviv, Accepted
[Paper]
HairFIT: Pose-invariant Hairstyle Transfer via Flow-based Hair Alignment and Semantic-region-aware Inpainting
Chaeyeon Chung*, Taewoo Kim*, Hyelin Nam*, Seunghwan Choi, Gyojung Gu, Sunghyun Park, and Jaegul Choo
British Machine Vision Conference (BMVC), 2021, Virtual, Accepted as Oral Presentation.
Best Paper Award, Korean Artificial Intelligence Association 2021.
[Paper]
Continuous-Time Video Generation via Learning Motion Dynamics with Neural ODE
Kangyeol Kim*, Sunghyun Park*, Junsoo Lee, Joonseok Lee, Sookyung Kim, Jaegul Choo, and Edward Choi
British Machine Vision Conference (BMVC), 2021, Virtual, Accepted.
[Paper]
K-Hairstyle: A Large-scale Korean Hairstyle Dataset for Virtual Hair Editing and Hairstyle Classification
Taewoo Kim*, Chaeyeon Chung*, Sunghyun Park*, Gyojung Gu, Keonmin Nam, Wonzo Choe, Jaesung Lee and Jaegul Choo
IEEE International Conference on Image Processing (ICIP), 2021, Virtual, Accepted.
[Paper]
[Project & Dataset]
[News]
VITON-HD: High-Resolution Virtual Try-On via Misalignment-Aware Normalization
Seunghwan Choi*, Sunghyun Park*, Minsoo Lee*, and Jaegul Choo
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021, Virtual, Accepted (27% acceptance rate).
[Paper]
[Project]
[Code & Dataset]
[PPT]
Vid-ODE: Continuous-Time Video Generation with Neural Ordinary Differential Equation
Sunghyun Park*, Kangyeol Kim*, Junsoo Lee, Jaegul Choo, Joonseok Lee, Sookyung Kim, and Edward Choi
AAAI Conference on Artificial Intelligence (AAAI), 2021, Virtual, Accepted (21% acceptance rate).
Qualcomm Innovation Fellowship Korea 2021.
[Paper]
[Project]
[Code & Dataset]
[PPT]
Hurricane Nowcasting with Irregular Time-step using Neural-ODE and Video Prediction
Sunghyun Park*, Kangyeol Kim*, Sookyung Kim*, Joonseok Lee, Junsoo Lee, Jiwoo Lee, and Jaegul Choo
International Conference on Learning Representations Workshop (ICLRW), 2020, Accepted as Spotlight Presentation.
[Paper]
PP-VTON: Pose-Preserving Image-based Virtual Try-On Network
Sunghyun Park* and Seunghwan Choi*
Korea Software Congress (KSC), 2019, Pyeongchang, Korea, Accepted.
[Paper]
Learning to Focus and Track Extreme Climate Events
Sookyung Kim*, Sunghyun Park*, Sunghyo Chung*, Joonseok Lee, Yunsung Lee, Hyojin Kim, Mr Prabhat, and Jaegul Choo
British Machine Vision Conference (BMVC), 2019, Cardiff, UK, Accepted as Spotlight Presentation (6.9% acceptance rate for spotlight papers).
ICML 2019 Workshop Climate Change, 2019, California, USA, Accepted.
NIPS 2019 Workshop Tackling Climate Change, 2019, Vancouver, Canada, Accepted.
[Paper]
[Aug. 2022 - Aug. 2023] Qualcomm AI Research, Research Intern (Morpheus Team) [Jul. 2021 - Dec. 2021] Kakao Enterprise, Membership Program (OCR Team) [Mar. 2021 - Jun. 2021] Kakao Enterprise, Research Intern (OCR Team) [Jul. 2018 - Aug. 2019] DAVIAN, Korea University, Undergraduate Researcher (Advised by Prof. Jaegul Choo) [Jan. 2018 - Feb. 2018] Daumsoft, Intern
Conference Reviewers: CVPR, ECCV, ICCV, NeurIPS, EuroGraphics, ICIP