Dongwoo Lee

Seonguk Seo

Hyewon Yoo

Seungmin Lee

Taehoon Kim

Tackgeun You

Minsoo Kang

Euyoung Kim

Sanghyun Son

Donghwan Jang

Jonathan Samuel Lumentut

Heewon Kim

Seon Jong Hyeon

Ilchae Jung

Sungyong Baik

Hongsuk Choi

Seungjun Nah

Jaedong Hwang

Gyeongsik Moon

Janghoon Choi

Gwangmo Song

Myungsub Choi

Geonwoon Jang

Junha Kim

Daniel Kim

Seohyun Kim

Jonghwan Mun

Paul Hongsuck Seo

Sunwoo Oh

Seungyeon Shin

Yumin Suh

Hyeonwoo Noh

Mohammad Amin Shabani

Jeany Son

Junhyung Kwak

Dongmin Park

Wonseok Lee

Woojae Chang

Mooyeol Baek

Heesoo Myeong

Haesol Park

Jungkwon Lee

Jungmin Lee

Tae Hyun Kim

Kamil Adamczewski

Sergei Liubich

Dong Woo Park

Young Min Shin

Wonsik Kim

Byeongjoo Ahn

Hee Seok Lee

Junseok Kwon

Junha Roh

Sangdon Park

Minsu Cho

Hyojin Kim

Hyo-Chan Son

Junghyun Kwon

Young Ki Baik

Stepan Tulyakov

Ivan Koryakovskiy

Moon Bin Lee

Jonghyun Choi

Joonyoung Park

Woo Yeon Jeong

Yoonkyung Kim

Hyun-Mook Cho

San Ko

Sung Won Park

Cheol Ki Hong

Kwang Soo Lim

Seung Hwan Choi

Won Ho Seo

Sang Woo Seok

Hyoung Jin Kim

Sung Min Jo

Jae Sung Han

Hyoung Lim Kim

Hanmin Ryu

Sang Hyup Lee

Yong Seok Heo

Tae Hoon Kim

Hoyub Jung

Soochahn Lee

Ju Yong Chang

Dong hyuk Lee

Yoon Phil Eo

Taeyoung Kim

Bogeon Park

HyunJun Oh

Jaechul Kim

The Computer Vision Laboratory, established at Seoul National University in 2003 under the supervision of Prof. Kyoung Mu Lee, conducts research and development in various areas of computer vision. The laboratory focuses on extracting, reconstructing, and recognizing information pertaining to objects, people, and 3D environments from digital images to advance artificial intelligence. Current research interests include deep learning, image restoration, image retrieval, scene understanding, object tracking/segmentation, human pose estimation, medical imaging, and 3D vision.

서울대학교 컴퓨터 비전 연구실은 2003년 Kyoung Mu Lee 교수의 지도하에 설립되었으며, 컴퓨터 비전 분야의 다양한 주제를 연구하고 개발한다. 인공 지능을 발전시키기 위해 디지털 이미지에서 객체, 사람 및 3D 환경 정보를 추출, 재구성 및 인식하는 데 중점을 둔다. 주요 연구 분야는 딥러닝, 이미지 복원, 이미지 검색, 장면 이해, 객체 추적/분할, 인간 자세 추정, 의료 영상 및 3D 비전이다.

Bridging the Distribution Gap to Harness Pretrained Diffusion Priors for Super-Resolution

LucidDreamer: Domain-free Generation of 3D Gaussian Splatting Scenes

Learning Dense Hand Contact Estimation from Imbalanced Data

Auto-regressive transformation for image alignment

PARTE: Part-Guided Texturing for 3D Human Reconstruction from a Single Image

Exploiting Diffusion Prior for Task-driven Image Restoration

MEIL-NeRF: Memory-Efficient Incremental Learning of Neural Radiance Fields

OmniSplat: Taming Feed-Forward 3D Gaussian Splatting for Omnidirectional Images with Editable Capabilities

DeClotH: Decomposable 3D Cloth and Human Body Reconstruction from a Single Image

SemanticDraw: Towards Real-Time Interactive Content Creation from Image Diffusion Models

Find A Winning Sign: Sign Is All We Need to Win the Lottery

GS-Blur: A 3D Scene-Based Dataset for Realistic Image Deblurring

ODGS: 3D Scene Reconstruction from Omnidirectional Images with 3D Gaussian Splattings

3D Hand Sequence Recovery from Real Blurry Images and Event Stream

CLOSER: Towards Better Representation Learning for Few-Shot Class-Incremental Learning

Overcoming Distribution Mismatch in Quantizing Image Super-Resolution Networks

Depth-Regularized Optimization for 3D Gaussian Splatting in Few-Shot Images

T2LM: Long-Term 3D Human Motion Generation from Multiple Sentences

Beyond Image Super-Resolution for Image Recognition with Task-Driven Perceptual Loss

CNC-Net: Self-Supervised Learning for CNC Machining Operations

AdaBM: On-the-Fly Adaptive Bit Mapping for Image Super-Resolution

Joint Reconstruction of 3D Human and Object via Contact-Based Refinement Transformer

ICF-SRSR: Invertible scale-Conditional Function for Self-Supervised Real-world Single Image Super-Resolution

3DHR-Co: A Collaborative Test-Time Refinement Framework for In-the-Wild 3D Human-Body Reconstruction Task

CoLaNet: Adaptive Context and Latent Information Blending for Face Image Inpainting

Abnormality Detection in Chest X-Ray via Residual-Saliency From Normal Generation

Extract-and-Adaptation Network for 3D Interacting Hand Mesh Recovery

ExBluRF: Efficient Radiance Fields for Extreme Motion Blurred Images

Cyclic Test-Time Adaptation on Monocular Video for 3D Human Mesh Reconstruction

Human Part-wise 3D Motion Context Learning for Sign Language Recognition

Content-Aware Local GAN for Photo-Realistic Super-Resolution

Dense Depth-Guided Generalizable NeRF

ACL-SPC: Adaptive Closed-Loop system for Self-Supervised Point Cloud Completion

Recovering 3D Hand Mesh Sequence from a Single Blurry Image: A New Dataset and Temporal Unfolding

NERDS: A General Framework to Train Camera Denoisers from Raw-RGB Noisy Image Pairs

Empirical Study of Pre-training a Backbone for 3D Human Pose and Shape Estimation

Long-Term 3D Human Motion Generation from Multiple Action Labels

Fine-Grained Neural Architecture Search for Image Super-Resolution

CADyQ: Content-Aware Dynamic Quantization for Image Super Resolution

3D Clothed Human Reconstruction in the Wild

Attentive Fine-Grained Structured Sparsity for Image Restoration

Learning to Estimate Robust 3D Human Mesh from In-the-Wild Crowded Scenes

AP-BSN: Self-Supervised Denoising for Real-World Images via Asymmetric PD and Blind-Spot Network

HandOccNet: Occlusion-Robust 3D Hand Mesh Estimation Network

CVF-SID: Cyclic multi-Variate Function for Self-Supervised Image Denoising by Disentangling Noise from Image

Map the Flow: Revealing Hidden Pathways of Information in VideoLLMs

FedLPA: Local Prior Alignment for Heterogeneous Federated Generalized Category Discovery

Fine-Grained Video Captioning through Scene Graph Consolidation

Enhanced Diffusion Sampling via Extrapolation with Multiple ODE Solutions

Revisiting Machine Unlearning with Dimensional Alignment

Metric Compatible Training for Online Backfilling in Large-Scale Retrieval

Re-evaluating Group Robustness via Adaptive Class-Specific Scaling

FIFO-Diffusion: Generating Infinite Videos from Text without Training

Hierarchical Visual Feature Aggregation for OCR-Free Document Understanding

4D Gaussian Splatting in the Wild with Uncertainty-Aware Regularization

Diffusion-Based Image-to-Image Translation by Noise Correction via Prompt Interpolation

Leveraging temporal contextualization for video action recognition

Robust Image Denoising through Adversarial Frequency Mixup

Communication-Efficient Federated Learning with Accelerated Client Gradient

Observation-Guided Diffusion Probabilistic Models

Relaxed Contrastive Learning for Federated Learning

Cross-Class Feature Augmentation for Class Incremental Learning

Randomized Adversarial Style Perturbations for Domain Generalization

Conditional Score Guidance for Text-Driven Image-to-Image Translation

Generative Neural Fields by Mixtures of Neural Implicit Functions

Beyond Pretrained Features: Noisy Image Modeling Provides Adversarial Defense

On the Stability-Plasticity Dilemma of Class-Incremental Learning

Multi-Modal Representation Learning with Text-Driven Soft Masks

Open-Set Representation Learning through Combinatorial Embedding

Variational Distribution Learning for Unsupervised Text-to-Image Generation

Stop or Forward: Dynamic Layer Skipping for Efficient Action Recognition

MCL-GAN: Generative Adversarial Networks with Multiple Specialized Discriminators

Information-Theoretic Generative Model Compression with Variational Energy-based Model

Locally Hierarchical Auto-Regressive Modeling for Image Generation

Towards Sequence-Level Training for Visual Tracking

Learning Semantic Segmentation from Multiple Datasets with Label Shifts

Multi-Level Branched Regularization for Federated Learning

Online Hybrid Lightweight Representations Learning: Its Application to Visual Tracking

Pooling Revisited: Your Receptive Field is Sub-optimal

Class-Incremental Learning by Knowledge Distillation with Adaptive Feature Consolidation

Unsupervised Learning of Debiased Representations with Pseudo-Attributes

InfoNeRF: Ray Entropy Minimization for Few-Shot Neural

Kyoung Mu Lee

Professor, Ph. D.

Learning Deconvolution Network for Semantic Segmentation

Learning multi-domain convolutional neural networks for visual tracking

The Visual Object Tracking VOT2015 Challenge Results

The Visual Object Tracking VOT2016 challenge results

Online Tracking by Learning Discriminative Saliency Map with Convolutional Neural Network

Large scale image retrieval with attentive deep local features

Domain-specific batch normalization for unsupervised domain adaptation

Weakly supervised action localization by sparse temporal pooling network

Image question answering using convolutional neural network with dynamic parameter prediction

Decoupled deep neural network for semi-supervised semantic segmentation

Modeling and propagating cnns in a tree structure for visual tracking

Channel attention is all you need for video frame interpolation

Local-global video-text interactions for temporal grounding

Real-time MDNet

Learning to optimize domain specific normalization for domain generalization

Multi-object tracking with quadruplet convolutional neural networks

Urvos: Unified referring video object segmentation network with a large-scale benchmark

Computer Vision Lab. Seoul National University

교수님 소개

이경무

연구실 소개

연구실 키워드

알럼나이가 간 회사

연구실 지원 방법

대우 조건

연도별 논문 인용수

저널 & 학회 논문 출판 수

연구실 논문 목록

Bridging the Distribution Gap to Harness Pretrained Diffusion Priors for Super-Resolution

Map the Flow: Revealing Hidden Pathways of Information in VideoLLMs

Bridging the Distribution Gap to Harness Pretrained Diffusion Priors for Super-Resolution

연구실 보유 특허

대학별 입시끈

우리 연구실 모집

교수님 소개

이경무

연구실 키워드

알럼나이가 간 회사

연구실 지원 방법

대우 조건

연도별 논문 인용수

저널 & 학회 논문 출판 수

연구실 논문 목록

Bridging the Distribution Gap to Harness Pretrained Diffusion Priors for Super-Resolution

Map the Flow: Revealing Hidden Pathways of Information in VideoLLMs

Bridging the Distribution Gap to Harness Pretrained Diffusion Priors for Super-Resolution

연구실 보유 특허

최근 3년간 연구실적 높은 연구실

대학별 입시끈

우리 연구실 모집