Kwan Yun

Kwan Yun (윤관)

Bridging 2D vision and 3D graphics to build human-centric generative models for controllable animation and high-fidelity synthesis

LinkedIn Scholar Github

News

  • MAR 2026: I will be joining Microsoft Research Cambridge as a research intern this summer.
  • AUG 2025: My research on face editing was featured in KAIST News.
  • JUNE 2025: I presented two first-authored papers at CVPR 2025.
  • July 2024: I received the Best Master’s Thesis Award from the Korea Computer Graphics Society

Selected Publications

StyleID: A Perception-Aware Dataset and Metric for Stylization-Agnostic Facial Identity Recognition

StyleID: A Perception-Aware Dataset and Metric for Stylization-Agnostic Facial Identity Recognition

Kwan Yun, Changmin Lee, AYeong Jeong, Youngseo Kim, Seungmi Lee, Junyong Noh
SIGGRAPH 2026; ACM Transactions on Graphics

Stylization-Agnostic Identity Encoder.

Project Page arXiv Code
While still early in his PhD journey, he has earned respect within Korea’s computer graphics and vision community through innovative work like AnyMoLe’s video diffusion approach for motion in-betweening and FFaceNeRF’s few-shot 3D face editing. His recognized as one of the most influential emerging researchers in the field of computer graphics and computer vision, specifically animation technology.
X-AVDT: Audio-Visual Cross-Attention for Robust Deepfake Detection

X-AVDT: Audio-Visual Cross-Attention for Robust Deepfake Detection

Youngseo Kim, Kwan Yun, , Sihun Cha, colette Koo, Junyong Noh
CVPR 2026

Diffusion cross-attention for deepfake detection.

Project Page arXiv Code
While still early in his PhD journey, he has earned respect within Korea’s computer graphics and vision community through innovative work like AnyMoLe’s video diffusion approach for motion in-betweening and FFaceNeRF’s few-shot 3D face editing. His recognized as one of the most influential emerging researchers in the field of computer graphics and computer vision, specifically animation technology.
StyleMM: Stylized 3D Morphable Face Model via Text-Driven Aligned Image Translation

StyleMM: Stylized 3D Morphable Face Model via Text-Driven Aligned Image Translation

Seungmi Lee*, Kwan Yun*, Junyong Noh
PG 2025; CGF

The first text generated stylized 3DMM.

Project Page arXiv Code
While still early in his PhD journey, he has earned respect within Korea’s computer graphics and vision community through innovative work like AnyMoLe’s video diffusion approach for motion in-betweening and FFaceNeRF’s few-shot 3D face editing. His recognized as one of the most influential emerging researchers in the field of computer graphics and computer vision, specifically animation technology.
AnyMoLe: Any Character Motion In-betweening Leveraging Video Diffusion Models

AnyMoLe: Any Character Motion In-betweening Leveraging Video Diffusion Models

Kwan Yun, Seokhyeon Hong, Chaelin Kim, Junyong Noh
CVPR 2025

AnyMoLe performs motion in-betweening for arbitrary characters only using inputs.

Project Page arXiv Code
While still early in his PhD journey, he has earned respect within Korea’s computer graphics and vision community through innovative work like AnyMoLe’s video diffusion approach for motion in-betweening and FFaceNeRF’s few-shot 3D face editing. His recognized as one of the most influential emerging researchers in the field of computer graphics and computer vision, specifically animation technology.
FFaceNeRF: Few-shot Face Editing in Neural Radiance Fields

FFaceNeRF: Few-shot Face Editing in Neural Radiance Fields

Kwan Yun, Chaelin Kim, Hangyeul Shin, Junyong Noh
CVPR 2025

Mask-based 3D face editing using a customized layout, trained with few-shot learning..

Project Page arXiv Code
While still early in his PhD journey, he has earned respect within Korea’s computer graphics and vision community through innovative work like AnyMoLe’s video diffusion approach for motion in-betweening and FFaceNeRF’s few-shot 3D face editing. His recognized as one of the most influential emerging researchers in the field of computer graphics and computer vision, specifically animation technology.
LeGO: Leveraging a Surface Deformation Network for Animatable Stylized Face Generation with One Example

LeGO: Leveraging a Surface Deformation Network for Animatable Stylized Face Generation with One Example

Soyeon Yoon*, Kwan Yun*, Kwanggyoon Seo, Sihun Cha, Jung Eun Yoo, Junyong Noh
CVPR 2024

Stylized and Animatable face mesh with one example.

Project Page arXiv
While still early in his PhD journey, he has earned respect within Korea’s computer graphics and vision community through innovative work like AnyMoLe’s video diffusion approach for motion in-betweening and FFaceNeRF’s few-shot 3D face editing. His recognized as one of the most influential emerging researchers in the field of computer graphics and computer vision, specifically animation technology.
Stylized Face Sketch Extraction via Generative Prior with Limited Data

Stylized Face Sketch Extraction via Generative Prior with Limited Data

Kwan Yun*, Kwanggyoon Seo*, Chang Wook Seo*, Soyeon Yoon, Seongcheol Kim, Soohyun Ji, Amirsaman Ashtari, Junyong Noh
EuroGraphics 2024; CGF

Utilizing the generative feature to extract and edit face sketch.

Project Page arXiv Code
While still early in his PhD journey, he has earned respect within Korea’s computer graphics and vision community through innovative work like AnyMoLe’s video diffusion approach for motion in-betweening and FFaceNeRF’s few-shot 3D face editing. His recognized as one of the most influential emerging researchers in the field of computer graphics and computer vision, specifically animation technology.