Research
I measure how far models trained on synthetic BIM data transfer to real buildings, using pre-registered experiments that fix predictions before execution.
Validity of Generative BIM Datasets as Training Data for Element Classification: Transfer Performance to Real-World BIM Data and Gap Factors
Jong Gwang Kim
대한건축학회 논문집
Jong Gwang Kim
CAADRIA 2027
A GNN-based Urban Context Feedback Framework for Early-Stage Architectural Design
Jong Gwang Kim, Hyeoncheol Kim
한국CDE학회 하계학술대회
Performance Comparison of Supervised and Self-Supervised Learning Graph Neural Networks for BIM Element Classification
Jong Gwang Kim, Hyeoncheol Kim
고려대학교