공지: 2026 후기 대학원생/학부 인턴을 모집합니다.
관련 정보
VSLab Visual Science Laboratory
Lab main visual 1
WELCOME TO
Visual Science Lab.
Our mission is to bridge the gap between human cognition and artificial intelligence — enabling machines to perceive the world more like humans do.
Application See Publications
Research highlight 1
ParTY: Part-Guidance for Expressive Text-to-Motion Synthesis,
CVPR 2026
Learn more
Research highlight 2
A Semantically Disentangled Unified Model for Multi-category 3D Anomaly Detection, CVPR 2026
Learn more
Research highlight 3
Object-Centric Representation Learning for Enhanced 3D Semantic Scene Graph Prediction, NeurIPS 2025
Learn more
Research highlight 4
Do We Need Perfect Data? Leveraging Noise for Domain Generalized Segmentation, AAAI 2026
Learn more
Research highlight 5
RefineVAD: Semantic-Guided Feature Recalibration for Weakly Supervised Video Anomaly Detection, AAAI 2026
Learn more
Research highlight 6
PlugTrack: Multi-Perceptive Motion Analysis for Adaptive Fusion in Multi-Object Tracking, AAAI 2026
Learn more

Announcement

“Our mission is to bridge the gap between human cognition and artificial intelligence, unraveling the complexities of how machines can perceive the world akin to humans.”

We are always looking for self-motivated and passionate PhD/MS students, Postdoc, and Undergraduate interns to join our lab. ( Application )

See more...

News

February 2026
Two papers led by SuYeon Kim, KunHo Heo have been accepted to CVPR 2026.
January 2026
Three papers led by Taeyeong Kim, Junhee Lee, and Seungjae Kim have been accepted to AAAI 2026.
September 2025
Our paper, KH Heo et al. "Object-Centric Representation Learning for Enhanced 3D Semantic Scene Graph Prediction," has been accepted to NeurIPS 2025.
September 2025
Myoungchul Kim has been awarded the NRF "Research Subsidies for Master’s Students" grant (Annual research budget: 12 million KRW).
August 2025
Our paper "Distance-based Bi-directional Edge Update Graph Neural Network for 3D Scene Graph Generation" has won the Excellence Award at the Junior Paper Competition from KCC 2025.