Regular biography
Seunghoon Hong is a faculty member in the School of Computing at KAIST. His research focuses on areas related to computer science, though specific details are not provided. He is affiliated with KAIST's Department of Computer Science and can be contacted via email at seunghoon.hong@kaist.ac.kr. His professional website is hosted at https://cvlab.postech.ac.kr/~maga33/.
Scholar-generated biography
Seunghoon Hong is an Associate Professor at KAIST with expertise in Computer Vision, Machine Learning, and Deep Learning. His research focuses on advancing semantic segmentation, video prediction, and generative modeling through deep neural networks. He has developed methods for weakly supervised segmentation, pseudo label refinement, and hierarchical text-to-image synthesis. His work also explores unsupervised learning, adversarial defense, and cross-domain disentanglement. Hong's contributions include learning transferrable knowledge, improving image clustering, and leveraging transformers for graph learning. His research emphasizes structured representations, diversity-sensitive generative models, and robust learning techniques for complex visual tasks.