Jeong-gi Kwak

I'm a postdoctoral research fellow at University of British Columbia (UBC), working with Prof. Kwang Moo Yi. My research interests are in image/video generative models and 3D computer vision, with the broader goal of modeling and interacting with the visual world. Before joining UBC, I worked in industry startups. At NXN Labs, I developed large-scale generative foundation models for fashion imagery, and at Innerverz AI, I focused on video diffusion models. I received my Ph.D. in Electrical Engineering from Korea University, where I was advised by Prof. Hanseok Ko. I also completed my B.S. and M.S. degrees in Electrical Engineering at Korea University.

Google Scholar  /  Github  /  CV  /  Email  /  Linkedin

News

  • Sep. 2025: One paper has been accepted to SIGGRAPH Asia 2025
  • Aug. 2025: Released Voost, a unified DiT achieving SOTA in virtual try-on and try-off.
  • Jul. 2025: In September, I will start a postdoctoral fellow position at the University of British Columbia (UBC), working with Prof. Kwang Moo Yi.
  • Jun. 2025: Recognized as an Outstanding Reviewer at CVPR 2025 (711/12,593 reviewers).
  • May 2025: One paper has been accepted to CVPRW 2025.
  • Dec. 2024: Joined NXN Labs as an AI researcher, focusing on visual generative/editing models.
  • Jul. 2024: I will be giving a talk at Twelve Labs.
  • Apr. 2024: Our paper has been selected as one of Highlight Papers at CVPR 2024 (Top 10%).
  • Feb. 2024: One paper has been accepted to CVPR 2024.
  • Jan. 2024: One paper has been accepted to ICASSP 2024.
  • Jan. 2024: Joined Innerverz AI as an AI/ML researcher, focusing on video diffusion models.
  • Dec. 2023: Successfully defended my thesis, “Towards Controllable and Interpretable Generative Neural Rendering”.
  • Dec. 2023: Completed a 6-month visit to the University of British Columbia (UBC) as a PhD visiting student in the Computer Vision Lab, supervised by Prof. Kwang Moo Yi.

Selected Publications

Voost: A Unified and Scalable Diffusion Transformer for Bidirectional Virtual
Try-On and Try-Off
Seungyong Lee*, Jeong-gi Kwak*
ACM SIGGRAPH Asia, 2025
paper | project page | demo
ViVid-1-to-3: Novel View Synthesis with Video Diffusion Models
Jeong-gi Kwak*, Erqun Dong*, Yuhe Jin, Hanseok Ko, Shweta Mahajan, Kwang Moo Yi
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024
Highlight (Top 10%)
paper | code | project page
Towards Multi-domain Face Landmark Detection with Synthetic Data from Diffusion Model
Yuanming Li, Gwantae Kim, Jeong-gi Kwak, Bonhwa Ku, Hanseok Ko
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024
paper
Injecting 3D Perception of Controllable NeRF-GAN into StyleGAN for Editable Portrait
Image Synthesis
Jeong-gi Kwak, Yuanming Li, Dongsik Yoon, Donghyeon Kim, David Han, Hanseok Ko
European Conference on Computer Vision (ECCV), 2022
ETNews ICT Paper Awards sponsored by MSIT Korea
paper | code | project page
DIFAI: Diverse Facial Inpainting using StyleGAN Inversion
Dongsik Yoon, Jeong-gi Kwak, Yuanming Li, David Han, Hanseok Ko
IEEE International Conference on Image Processing (ICIP), 2022
paper
Generate and Edit Your Own Character in a Canonical View
Jeong-gi Kwak, Yuanming Li, Dongsik Yoon, David Han, Hanseok Ko
IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2022
paper | poster
Adverse Weather Image Translation with Asymmetric and Uncertainty-aware GAN
Jeong-gi Kwak, Youngsaeng Jin, Yuanming Li, Dongsik Yoon, Donghyeon Kim, Hanseok Ko
British Machine Vision Conference (BMVC), 2021
paper | code
Reference Guided Image Inpainting using Facial Attributes
Dongsik Yoon, Jeong-gi Kwak, Yuanming Li, David Han, Youngsaeng Jin, Hanseok Ko
British Machine Vision Conference (BMVC), 2021
paper | code
CAFE-GAN: Arbitrary Face Attribute Editing with Complementary Attention Feature
Jeong-gi Kwak, David K. Han, Hanseok Ko
European Conference on Computer Vision (ECCV), 2020
paper

Academic Service

Reviewer for international conferences and journals:
  • CVPR (2025, 2024, 2022)
  • NeurIPS (2025, 2023)
  • AAAI (2025)
  • SIGGRAPH Asia (2025)
  • WACV (2026)
  • ECCV (2022)
  • ICASSP (2023, 2022, 2021)
  • CVIU (2023)
This website template is adapted from Jon Barron's homepage .