Youngseo Kim
I am a MS student at KAIST, advised by professor Junyoung Noh in the Visual Media Lab. My research focuses on Generative Models, with a particular interest in areas such as deepfake detection, NSFW content moderation, copyright protection, and privacy-preserving AI systems.
Latest News
- APRIL 2026: 1 Paper accepted to SIGGRAPH 2026
- FEB 2026: 1 Paper accepted to CVPR 2026
Selected Publications
Perception-aware dataset and metric for facial identity recognition that remains robust under stylized images, supporting identity similarity measurement, retrieval, and evaluation.
Robust deepfake detection method and cross-generator dataset by leveraging cross-attention features to capture audio-visual correlations.
Stable Diffusion feature extraction for sketching with one example, enabling diverse stylized sketch generation and efficient distilled sketch extraction.