Digital Avatar Lab

Sungshin Women's University

Computer Graphics

Computer Vision

Artificial Intelligence

Our lab conducts cutting-edge research at the intersection of Computer Graphics, Computer Vision, and Artificial Intelligence. We focus on developing next-generation visual computing technologies that bridge the physical and digital worlds, enabling realistic and immersive digital experiences.

Our research centers on AI-based digital avatar creation, including expressive facial animation and high-fidelity virtual character generation. By integrating deep learning with advanced graphics techniques, we aim to create realistic and natural digital avatars. Through this work, we advance digital human technology and contribute to the future of intelligent visual media.

📢 Join Us

We are looking for highly motivated and passionate students and researchers who are excited about computer graphics, computer vision, and AI. If you are enthusiastic about building the future of digital humans, immersive media, and AI-driven graphics technologies, we would love to hear from you. 💌Please contact: sunjin@sungshin.ac.kr

Let's shape the future of digital avatar technology and visual computing together ✨

Professor

Sunjin Jung, Ph.D.

Assistant Professor
Department of Computer Engineering,
Sungshin Women's University

sunjin@sungshin.ac.kr

Google Scholar

Education

Ph.D. in Graduate School of Culture Technology, KAIST, 2024

Thesis: Audio-Driven 3D Facial Animation Generation
Advisor: Prof. Junyong Noh

M.S. in Graduate School of Culture Technology, KAIST, 2017

Thesis: Age-related Gait Motion Generation Based on Biomechanical Observations
Advisor: Prof. Junyong Noh

B.S. in Department of Digital Media, Ajou University, 2015

Work Experience

Assistant Professor, Sungshin Women's University, 2025.03 ~ Present
Postdoctoral Researcher, Visual Media Lab, KAIST, 2024.09 ~ 2025.02
Research Intern, NAVER Cloud, 2023.10 ~ 2024.01

Projects

Generating Hyper-Realistic to Extremely-stylized Face Avatar with Varied Speech Speed and Context-based Emotional Expression

IITP, 2024.07 ~ 2025.02

Development of Virtual Space Automation and Intelligent Editor Technology based on Space Interface

KOCCA, 2024.07 ~ 2025.02

Development of universal fashion creation platform technologh for avatar personality expression

KOCCA, 2023.04 ~ 2024.08

Development and commercialization of user generation tool platform for production of realistic 3D emoji interaction based on artificial intelligence meta technology

KOCCA, 2023.04 ~ 2024.03

Development of self-evolving AI creation platform

KOCCA, 2020.07 ~ 2022.12

A Deep Learning Based Immersive AR Content Creation Platform for Generating Interactive, Context and Geometry Aware Movement from a Single Image

IITP, 2020.04 ~ 2022.12

Emotional Intelligence Technology to Infer Human Emotion and Carry on Dialogue Accordingly

IITP, 2016.12 ~ 2020.12

Teaching

Undergraduate Courses

Java Programming (Spring 2025, Spring 2026)
Web Programming Basics (Spring 2025, Spring 2026)
Introduction to Information Technology (Spring 2025, Spring 2026)
Mobile Programming (Fall 2025)
Mathematics for Digital Systems (Fall 2025)
Special Topics in Computer Engineering (Fall 2025)

Graduate Courses

Topics in Computer Graphics (Spring 2026)

Academic Services

Conference Committees

HCI Korea 2026 - Organizing Committee Member (Workshop and Emerging Scholars Session)
SIGGRAPH Asia 2025 - Technical Communications and Posters Committee Member

Journal Reviewer

ACM Transactions on Graphics (TOG)
IEEE Transactions on Visualization and Computer Graphics (TVCG)
Computer Graphics Forum (CGF)

Conference Reviewer

SIGGRAPH (2026, 2025)
SIGGRAPH Asia (2025)
Pacific Conference on Computer Graphics and Applications (PG) (2024)
International Conference on Computer Animation and Social Agents (CASA) (2023, 2022)
Computer Graphics International (CGI) (2020)
International Conference on Image and Vision Computing New Zealand (IVCNZ) (2019, 2018, 2017)
International Symposium on Electronic Art (ISEA) (2019)
Korea Computer Graphics Society (KCGS) (2020)
HCI Korea (2026, 2018)

Awards

Best Paper Award, Pacific Graphics (PG), 2024
Best Paper Award, Korea Computer Graphics Society (KCGS), 2020
Master's Thesis Award, Korea Computer Graphics Society (KCGS), 2017

Members

Sun Cho

B.S. Student | Undergraduate Researcher

20231045@sungshin.ac.kr

Jinseon Yang

B.S. Student | Undergraduate Researcher

20231145@sungshin.ac.kr

🌜Head Student🌛

Publications

Deep-Learning-Based Facial Retargeting Using Local Patches

Computer Graphics Forum (CGF)
Eurographics 2025

Speed-Aware Audio-Driven Speech Animation using Adaptive Windows

ACM Transactions on Graphics (TOG) SIGGRAPH Asia 2024

Audio-Driven Speech Animation with Text-Guided Expression

Pacific Graphics (PG) 2024
🏆 Best Paper Award

Deep Learning-Based Unsupervised Human Facial Retargeting

Computer Graphics Forum (CGF)
Pacific Graphics 2021

Speech Animation Synthesis based on a Korean Co-articulation Model

Korea Computer Graphics Society (KCGS) 2020 🏆 Best Paper Award

Synthesizing Character Animation with Smoothly Decomposed Motion Layers

Computer Graphics Forum (CGF)
Eurographics 2020

Age-related Gait Motion Transformation Based on Biomechanical Observations

Computer Animation and Virtual Worlds (CAVW), CASA 2017

SketchiMo: Sketch-based Motion Editing for Articulated Characters

ACM Transactions on Graphics (TOG) SIGGRAPH 2016

Abstract

In the era of digital animation, the quest to produce lifelike facial animations for virtual characters has led to the development of various retargeting methods. While the retargeting facial motion between models of similar shapes has been very successful, challenges arise when the retargeting is performed on stylized or exaggerated 3D characters that deviate significantly from human facial structures. In this scenario, it is important to consider the target character’s facial structure and possible range of motion to preserve the semantics assumed by the original facial motions after the retargeting. To achieve this, we propose a local patch-based retargeting method that transfers facial animations captured in a source performance video to a target stylized 3D character. Our method consists of three modules. The Automatic Patch Extraction Module extracts local patches from the source video frame. These patches are processed through the Reenactment Module to generate correspondingly re-enacted target local patches. The Weight Estimation Module calculates the animation parameters for the target character at every frame for the creation of a complete facial animation sequence. Extensive experiments demonstrate that our method can successfully transfer the semantic meaning of source facial expressions to stylized characters with considerable variations in facial feature proportion.

Abstract

We present a novel method that can generate realistic speech animations of a 3D face from audio using multiple adaptive windows. In contrast to previous studies that use a fixed size audio window, our method accepts an adaptive audio window as input, reflecting the audio speaking rate to use consistent phonemic information. Our system consists of three parts. First, the speaking rate is estimated from the input audio using a neural network trained in a self-supervised manner. Second, the appropriate window size that encloses the audio features is predicted adaptively based on the estimated speaking rate. Another key element lies in the use of multiple audio windows of different sizes as input to the animation generator: a small window to concentrate on detailed information and a large window to consider broad phonemic information near the center frame. Finally, the speech animation is generated from the multiple adaptive audio windows. Our method can generate realistic speech animations from inthe-wild audios at any speaking rate, i.e., fast raps, slow songs, as well as normal speech. We demonstrate via extensive quantitative and qualitative evaluations including a user study that our method outperforms state-ofthe-art approaches.

Abstract

We introduce a novel method for generating expressive speech animations of a 3D face, driven by both audio and text descriptions. Many previous approaches focused on generating facial expressions using pre-defined emotion categories. In contrast, our method is capable of generating facial expressions from text descriptions unseen during training, without limitations to specific emotion classes. Our system employs a two-stage approach. In the first stage, an auto-encoder is trained to disentangle content and expression features from facial animations. In the second stage, two transformer-based networks predict the content and expression features from audio and text inputs, respectively. These features are then passed to the decoder of the pre-trained auto-encoder, yielding the final expressive speech animation. By accommodating diverse forms of natural language, such as emotion words or detailed facial expression descriptions, our method offers an intuitive and versatile way to generate expressive speech animations. Extensive quantitative and qualitative evaluations, including a user study, demonstrate that our method can produce natural expressive speech animations that correspond to the input audio and text descriptions.