Research Area [ English | 한국어 ]
GIST Machine Learning and Vision Laboratory seeks to advance 1️⃣ computer vision and pattern recognition algorithms, particularly in the area of multi-object tracking and 2D / 3D visual perception (e.g. detection / action recognition / density estimation) using traditional algorithmic and recent machine learning approaches. We are interested in applications for real-time 2️⃣ visual surveillance. Besides, to help numerous IT companies use their AI models more efficiently, we also study various technologies such as knowledge distillation, model compression and imbalanced learning etc. Recently, 3️⃣ cultural technology has been in the spotlight and we try to contribute to the cultural industry with our ML/CV knowledge (e.g. exhibition hall guide robot) through cooperation with other domain experts.
Computer Vision & Pattern Recognition
(High-level & Low-level) Image / Video Processing
Representation Learning
Machine Learning
Model Compression
Knowledge Distillation
Imbalanced Learning
Visual Surveillance
Multi-object Tracking
2D / 3D Object Detection
Anomaly Detection
Density Estimation
Action Recognition
Re-identification
Cultural Technology
Exhibition Hall Guide Robot