Sona Sim

Gi Yun Choi

MinGyu Lee

Yeonghun Yoon

Taeheon Lee

Haeun Noh

Jaehyung Bae

Kanghee Lee

Cho Hyun Jin

Dohee Kim

Minji Kwak

Taehoon Kim

Suk Hyun Kim

Jooyoung Lee

Suyeon Cha

Yujeong Oh

Baek Seo Hyun

Soo Hyun Park

Jisu Park

Lim Hyoung Jun

Cho Ye-Eun

EunGyeong Bang

Jongwook Choi

PyoungGeon Kim

Hwang Jin Soo

JaeYoon Lee

Youn Jong Su

Jong Min Lee

Hojoon Jung

Heegon Jin

Young Wook Kang

Jun Sup Shin

Sejong Pyo

Seo Seung Mo

Seungjin Jung

Doyeon Kim

Yonghyun Jeong

The Visual Intelligence Lab (VILab) at Chung-Ang University (CAU) is part of the Graduate School of Advanced Imaging Science, which was established in 1999 to fuse visual arts and engineering. VILab focuses on artificial intelligence (AI) research, particularly in computer vision and deep learning, with the goal of developing new AI models and enhancing the performance of existing ones. The lab provides a systematic research environment, supporting students through R&D projects, scholarships, and dedicated GPU servers (RTX3090 for Master's, A6000 for Ph.D., and shared A100/A6000/Pro6000 with approximately 30 GPUs). It encourages diverse research topics and offers flexible work arrangements. VILab has a strong publication record in top-tier AI and computer vision conferences and journals.

중앙대학교 시각 지능 연구실(VILab)은 시각 예술과 공학의 융합을 위해 1999년에 설립된 첨단영상대학원 소속입니다. VILab은 새로운 AI 모델을 개발하고 기존 모델의 성능을 향상시키는 것을 목표로 컴퓨터 비전 및 딥러닝 분야의 인공지능(AI) 연구에 중점을 둡니다. 연구실은 R&D 프로젝트, 장학금, 전용 GPU 서버(석사 RTX3090, 박사 A6000, 공유 A100/A6000/Pro6000 약 30개 GPU)를 통해 학생들을 지원하며 체계적인 연구 환경을 제공합니다. 다양한 연구 주제를 장려하고 유연한 근무 방식을 제공합니다. VILab은 최고 수준의 AI 및 컴퓨터 비전 학회 및 저널에 강력한 출판 기록을 보유하고 있습니다.

Through the Water: Refractive Gaussian Splatting for Water Surface Scenes

Beyond Spatial Frequency: Pixel-wise Temporal Frequency-based Deepfake Video Detection

AJAHR: Amputated Joint Aware 3D Human Mesh Recovery

Group-wise Scaling and Orthogonal Decomposition for Domain-Invariant Feature Extraction in Face Anti-Spoofing

InsideOut: Integrated RGB-Radiative Gaussian Splatting for Comprehensive 3D Object Representation

NBA3D: Neighbor-Based Confidence Adjustment for 3D Rare Object Detection Using LiDAR

Exploiting Style Latent Flows for Generalizing Video Deepfake Detection

Text-Guided Variational Image Generation for Industrial Anomaly Detection and Segmentation

Scaling of Class-wise Training Losses for Post-hoc Calibration

FingerprintNet: Synthesized Fingerprints for Generated Image Detection

FrePGAN: Robust Deepfake Detection Using Frequency-level Perturbations

Domain-generalizable Face Anti-Spoofing with Patch-based Multi-tasking and Artifact Pattern Conversion

Online Adaptive Slimmable Network for Source-free Unsupervised Domain Adaptation

ID-RAG: Industrial Defect-Retrieval Augmented Generation for Robust and Precise Industrial Surface Defect Detection

3D-Aware Virtual Try-On Using Only 2D Inputs

Toward enhanced unsupervised clustering of 20th century Korean paintings via multimodal features

Explainable machine learning-based classification of traditional Korean ceramics using XRF chemical composition data

Style Prompt Tuning for Bridging Visual Gaps in Autonomous Driving

RICH: Coin data Restoration Using Inpainting for Cultural Heritage Analysis

Generating Visual-Adaptive Audio Representation for Audio Recognition

Multitask Adaptation for Unlabeled Domain Using Multiple Single-task Domains

Observations on K-image Expansion of Image-Mixing Augmentation

Active Instance Selection for Few-Shot Classification

Flexible Knowledge Distillation for Class-Incremental Learning via Structural Knowledge Transfer

Dynamic Augmentation Strategy Selection for Incremental Object Detection

Mitigating Class Imbalance in Neural Network Quantization via Rank-Based Variance Regularization

TPRNet: Texture PReserving Network for Realistic Scenery Image Extrapolation

Category-specific Augmentation Search Using Attentional Interpolation for Unsupervised Domain Adaptation

Adaptive Attention Link-based Regularization for Vision Transformers

Facemover: Unsupervised face frontalization with embedding segments

Few-shot Semantic Segmentation with Uncertainty-based Joint Prototypes

Detection of Videos with Audio-Visual Inconsistency for Video Representation Learning

MAYA: Multi-Attack Yielding Augmentation for Unified Face Attack Detection

Visual Design from Cultural Heritage through Inpainting-based Restoration

3D Gaussian Splatting for Enhanced Documentation of Cultural Artifacts

FH-RestoreASR: Frequency-Hopping Robust Air Traffic Control Speech Restoration and Recognition

SCENEDERELLA: Text-Driven 3D Scene Generation for Realistic Artboard

ID-Flow: Leveraging Voice Identity for Generalizing Audio Deepfake Detection

Topic-VQ-VAE: Leveraging Latent Codebooks for Flexible Topic-Guided Document Generation

Educational Research Trends of the Use of Gaze Learning Data through Topic Modeling and Scientometric Analysis

Document Change Detection with Hierarchical Patch Comparison

Domain Generalization for Face Forgery Detection by Style Transfer

BiHPF: Bilateral High-Pass Filters for Robust Detection of Synthesized Images

mToFNet: Object Anti-Spoofing with Mobile Time-of-Flight Data

Novel-View Synthesis of Human Tourist Photos

Tracking Failure Prediction for Siamese Trackers Based on Channel Feature Statistics

VaB-AL: Incorporating Class Imbalance and Difficulty with Variational Bayes for Active Learning

Visual Domain Adaptation by Consensus-based Transfer to Intermediate Domain

Associative Variational Auto-encoder with Distributed Latent Spaces and Associators

DoFNet: Depth of Field Difference Learning for Detecting Image Forgery

Backbone Can Not be Trained at Once: Rolling Back to Pre-trained Network for Person Re-Identification

Accelerating Simultaneous Optimization of the Model Architecture and the Training Hyperparameter

Context-aware Deep Feature Compression for High-speed Visual Tracking

Selective Ensemble Network for Accurate Crowd Density Estimation

Attentional Correlation Filter Network for Adaptive Visual Tracking

Action-Decision Network for Visual Tracking with Deep Reinforcement Learning

Effect of Learning Rate to Generalization Power

An Integrated Framework for 24-hours Fire Detection

Density-aware Pedestrian Proposal Networks for Robust People Detection in Crowded Scenes

Visual Tracking Using Attention-Modulated Disintegration and Integration

Patch-based Fire Detection with Online Outlier Learning

Robust Pan-Tilt-Zoom Tracking via Optimization Combining Motion Features and Appearance Correlations

User Interactive Segmentation with Partially Growing Random Forest

Gradient Preserving RGB-to-Gray Conversion using Random Forest

Example-based Denoising for Detail Texture

Scene-invariant Crowd Density Estimation with Aggregated Features

Balanced Optical Flow Refinement by Bidirectional Constraint

Single-camera Based Vehicle Pose Estimation Using Multiple Features on the Road Surface

Crosswalk and Traffic Light Detection via Integral Framework

Implementation of Driverless Car: Efficient Sensor Network for High Speed Robot

A Superpixel MRF Approach Using High-order Likelihood for Moving Object Detection

Self-supervised Scheme for Generalizing GAN Image Detection

Geographical discrimination of Asian red pepper powders using 1H-NMR spectroscopy and deep learning-based convolution neural networks

Active Instance Selection for Few-shot Classification

Optimization-Based Method Removing Highlights with Flash/No-Flash Image

Congestion-aware Bayesian Loss for Crowd Counting

Visual content analysis of visitors’ engagement with an instagrammable exhibition

Rollback Ensemble with Multiple Local Minima in Fine-tuning Deep Learning Networks

VILab@CAU

교수님 소개

최종원

연구실 소개

연구실 키워드

알럼나이가 간 회사

연구실 지원 방법

대우 조건

연도별 논문 인용수

저널 & 학회 논문 출판 수

연구실 논문 목록

Through the Water: Refractive Gaussian Splatting for Water Surface Scenes

Domain-generalizable Face Anti-Spoofing with Patch-based Multi-tasking and Artifact Pattern Conversion

Online Adaptive Slimmable Network for Source-free Unsupervised Domain Adaptation

연구실 보유 특허

대학별 입시끈

우리 연구실 모집

교수님 소개

최종원

연구실 키워드

알럼나이가 간 회사

연구실 지원 방법

대우 조건

연도별 논문 인용수

저널 & 학회 논문 출판 수

연구실 논문 목록

Through the Water: Refractive Gaussian Splatting for Water Surface Scenes

Domain-generalizable Face Anti-Spoofing with Patch-based Multi-tasking and Artifact Pattern Conversion

Online Adaptive Slimmable Network for Source-free Unsupervised Domain Adaptation

연구실 보유 특허

최근 3년간 연구실적 높은 연구실

대학별 입시끈

우리 연구실 모집