Daehyun Bae

Se-Yun Lee

Jaehyun Choi

Wonyoung Joo

Jaehyun Lew

Seungyeol Lee

Euiyeon Kim

Seunggu Kang

Sang-Heon Shim

Jihwan Kim

Suho Park

Visual Computing Laboratory at Sungkyunkwan University (VCLab@SKKU) conducts research in Computer Vision and Machine Learning. We are looking for highly motivated undergraduate, M.S., and Ph. D. students who want to enjoy the cutting-edge research. Please contact Jae-Pil Heo if you are interested in research in Computer Vision, Machine Learning, or other related areas.

성균관대학교 비주얼 컴퓨팅 연구실(VCLab@SKKU)은 컴퓨터 비전 및 머신러닝 분야를 연구합니다. 최첨단 연구를 즐기고 싶은 의욕 넘치는 학부생, 석사 및 박사 학생을 찾습니다. 컴퓨터 비전, 머신러닝 또는 기타 관련 분야 연구에 관심이 있다면 허재필 교수에게 문의하십시오.

Scalable GANs with Transformers

Masking Matters: Unlocking the Spatial Reasoning Capabilities of LLMs for 3D Scene-Language Understanding

SeaCache: Spectral-Evolution-Aware Cache for Accelerating Diffusion Models

Reconstruction-Guided Slot Curriculum: Addressing Object Over-Fragmentation in Video Object-Centric Learning

Looking Beyond the Window: Global-Local Aligned CLIP for Training-free Open-Vocabulary Semantic Segmentation

Temporally Consistent Long-Term Memory for 3D Single Object Tracking

PDF-GS: Progressive Distractor Filtering for Robust 3D Gaussian Splatting

Analyzing the Training Dynamics of Image Restoration Transformers: A Revisit to Layer Normalization

From Vicious to Virtuous Cycles: Synergistic Representation Learning for Unsupervised Video Object-Centric Learning

Diffusion Feature Field for Text-based 3D Editing with Gaussian Splatting

Mitigating Semantic Collapse in Partially Relevant Video Retrieval

Selective Contrastive Learning for Weakly Supervised Affordance Grounding

Prototypes are Balanced Units for Efficient and Effective Partially Relevant Video Retrieval

Fine-Tuning Visual Autoregressive Models for Subject-Driven Generation

Translation of Text Embedding via Delta Vector to Suppress Strongly Entangled Content in Text-to-Image Diffusion Models

Auto-Encoded Supervision for Perceptual Image Super-Resolution

Temporal Alignment-Free Video Matching for Few-shot Action Recognition

Prediction-Feedback DETR for Temporal Action Detection

Ambiguity-Restrained Text-Video Representation Learning for Partially Relevant Video Retrieval

Bridging the Semantic Granularity Gap Between Text and Frame Representations for Partially Relevant Video Retrieval

Foreground-Covering Prototype Generation and Matching for SAM-Aided Few-Shot Segmentation

GSGAN: Adversarial Learning for Hierarchical Generation of 3D Gaussian Splats

Activating Self-Attention for Multi-Scene Absolute Pose Regression

Progressive Proxy Anchor Propagation for Unsupervised Semantic Segmentation

Mitigating Background Shift in Class-Incremental Semantic Segmentation

Style Injection in Diffusion: A Training-free Approach for Adapting Large-scale Diffusion Models for Style Transfer

Diversity-aware Channel Pruning for StyleGAN Compression

Noise-free Optimization in Early Training Steps for Image Super-Resolution

Towards Squeezing-Averse Virtual Try-On via Sequential Deformation

Task-disruptive Background Suppression for Few-Shot Segmentation

VLCounter: Text-aware Visual Representation for Zero-Shot Object Counting

Self-Feedback DETR for Temporal Action Detection

Disentangled Representation Learning for Unsupervised Neural Quantization

Query-Dependent Video Representation for Moment Retrieval and Highlight Detection

Leveraging Hidden Positives for Unsupervised Semantic Segmentation

Progressive Few-shot Adaption of Generative Model with Align-free Spatial Correlation

Minority-Oriented Vicinity Expansion with Attentive Aggregation for Video Long-Tailed Recognition

Robust Image Denoising of No-Flash Images Guided by Consistent Flash Images

Tailoring Self-Supervision for Supervised Learning

Difficulty-Aware Simulator for Open Set Recognition

Local Attention Pyramid for Scene Image Generation

Task Discrepancy Maximization for Fine-grained Few-Shot Classification

Product Quantizer Aware Inverted Index for Scalable Nearest Neighbor Search

Collaborative Learning with Disentangled Features for Zero-shot Domain Adaptation

Self-Supervised Video GANs: Learning for Appearance Consistency and Motion Coherency

VarSR: Variational Super-Resolution Network for Very Low Resolution Images

Accurate and Consistent Depth Estimation for Light Field Camera Arrays

Plenoptic Imaging Techniques for Improving Accuracy and Robustness of Object Tracking

Shortlist Selection with Residual-Aware Distance Estimator for K-Nearest Neighbor Search

Distance Encoded Product Quantization

VLSH: Voronoi-based Locality Sensitive Hashing

Spherical Hashing

Quadra-Embedding: Binary Code Embedding with Low Quantization Error

IRIW: Image Retrieval based Image Watermarking for Large-Scale Image Databases

Data Management for SSDs for Large-Scale Interactive Graphics Applications

FASTCD: Fracturing-Aware Stable Collision Detection

Boundary-Recovering Network for Temporal Action Detection

Correlation-Guided Calibration of Query Dependency for Video Temporal Grounding

Task-Oriented Channel Attention for Fine-Grained Few-Shot Classification

Mutually-Aware Feature Learning for Few-Shot Object Counting

Detection of Infill Development and Contributing Factors using Deep Learning and Multilevel Modeling

Factors associated with pedestrian-vehicle collision hotspots involving seniors and children: a deep learning analysis of street-level images

TCX: Texture and Channel Swappings for Domain Generalization

The detection of residential developments in urban areas: Exploring the potentials of deep-learning algorithms

Frequency-based Motion Representation for Video Generative Adversarial Networks

Cross-Loss Pseudo Labeling for Semi-Supervised Segmentation

Axial Constraints for Global Matching-Based Optical Flow Estimation

Pivot-Guided Embedding for Domain Generalization

Pixel-Level Kernel Estimation for Blind Super-Resolution

Mutually Orthogonal Softmax Axes for Cross-Domain Retrieval

Pseudo Conditional Regularization for Inverse Mapping of GANs

Distance Encoded Product Quantization for Approximate K-Nearest Neighbor Search in High-Dimensional Space

Learning Coarse and Fine Features for Precise Temporal Action Localization

Content-Aware Focal Plane Selection and Proposals for Object Tracking on Plenoptic Image Sequences

Visual Computing Lab @ SKKU

교수님 소개

허재필

연구실 소개

연구실 키워드

알럼나이가 간 회사

연구실 지원 방법

대우 조건

연도별 논문 인용수

저널 & 학회 논문 출판 수

연구실 논문 목록

Scalable GANs with Transformers

Masking Matters: Unlocking the Spatial Reasoning Capabilities of LLMs for 3D Scene-Language Understanding

SeaCache: Spectral-Evolution-Aware Cache for Accelerating Diffusion Models

연구실 보유 특허

대학별 입시끈

우리 연구실 모집

교수님 소개

허재필

연구실 키워드

알럼나이가 간 회사

연구실 지원 방법

대우 조건

연도별 논문 인용수

저널 & 학회 논문 출판 수

연구실 논문 목록

Scalable GANs with Transformers

Masking Matters: Unlocking the Spatial Reasoning Capabilities of LLMs for 3D Scene-Language Understanding

SeaCache: Spectral-Evolution-Aware Cache for Accelerating Diffusion Models

연구실 보유 특허

최근 3년간 연구실적 높은 연구실

대학별 입시끈

우리 연구실 모집