Susung Hong

Welcome to my website! I am a Ph.D. student in GRAIL at the University of Washington, co-advised by Steve Seitz, Brian Curless, and Ira Kemelmacher-Shlizerman. My research interests lie in the foundations of generative AI, and its applications to producing immersive video/3D content.

I love cats, coffee (especially lattes), video games, cooking, and playing baseball. I enjoy learning new things beyond my research area.

If you are interested in collaboration or would like to learn more about my research, please connect:

Susung Hong

News

I’ve joined Google as a PhD Student Researcher.

MusicInfuser has been accepted to CVPR 2026.

Our two papers, Perturb-and-Revise and Spatiotemporal Skip Guidance (STG) have been accepted to CVPR 2025.

Our two papers, Smoothed Energy Guidance (SEG) and Effective Rank Regularization for 3DGS have been accepted to NeurIPS 2024.

Selected Publications

COMIC: Agentic Sketch Comedy Generation

COMIC: Agentic Sketch Comedy Generation

Preprint 2026

Susung Hong , Brian Curless , Ira Kemelmacher-Shlizerman , Steve Seitz

MusicInfuser: Making Video Diffusion Listen and Dance

MusicInfuser: Making Video Diffusion Listen and Dance

CVPR 2026

Susung Hong , Ira Kemelmacher-Shlizerman , Brian Curless , Steven M Seitz

Perturb-and-Revise: Flexible 3D Editing with Generative Trajectories

Perturb-and-Revise: Flexible 3D Editing with Generative Trajectories

CVPR 2025

Susung Hong , Johanna Karras , Ricardo Martin-Brualla , Ira Kemelmacher-Shlizerman

Spatiotemporal Skip Guidance for Enhanced Video Diffusion Sampling

Spatiotemporal Skip Guidance for Enhanced Video Diffusion Sampling

CVPR 2025

Junha Hyung* , Kinam Kim* , Susung Hong , Min-Jung Kim , Jaegul Choo

Smoothed Energy Guidance: Guiding Diffusion Models with Reduced Energy Curvature of Attention

Smoothed Energy Guidance: Guiding Diffusion Models with Reduced Energy Curvature of Attention

NeurIPS 2024

Susung Hong

Improving Sample Quality of Diffusion Models Using Self-Attention Guidance

Improving Sample Quality of Diffusion Models Using Self-Attention Guidance

ICCV 2023

Susung Hong , Gyuseong Lee , Wooseok Jang , Seungryong Kim

View all publications →

Experience

Google | PhD Student Researcher

Generative media team

Adobe | Research Scientist Intern

Video diffusion previews (Mentors: Jui-Hsien Wang, Chongjian Ge, Zhifei Zhang)

UW GRAIL | Research Assistant

Immersive video/3D content production (Advisors: Steve Seitz, Brian Curless, Ira Kemelmacher-Shlizerman)

NAVER | Research Intern

Effective rank regularization for 3DGS and generative relighting (Mentor: Jin-Hwa Kim)

UW GRAIL | Undergraduate Research Assistant

Flexible 3D editing with generative trajectories (Advisor: Ira Kemelmacher-Shlizerman, Ricardo Martin-Brualla, Johanna Karras)

Twelve Labs | ML Research Intern

Video-language model agents and high-quality dataset (Mentor: Minjoon Seo)

CVLAB | Undergraduate Research Assistant

Image, 3D, video generation, and neural fields (Advisor: Seungryong Kim)