김수필
Kim, SoopilDivision of Intelligent Robotics
학력
- 2019 ~ 2024대구경북과학기술원 석박통합과정
- 2014 ~ 2019대구경북과학기술원 학사
경력
- 2024 ~ 2025대구경북과학기술원 / 박사후연수연구원
수상실적
연구실 소개
- AI-Vision Lab
-
Our lab aims to address computer vision challenges across diverse domains—including natural images, medical imaging, and industrial scenes—by leveraging state-of-the-art AI technologies. We are dedicated to designing novel models, extracting meaningful information from data, and conducting in-depth analysis to solve real-world problems. Through these efforts, we strive to build intelligent systems that can robustly interpret complex visual environments and contribute to both scientific advancement and societal impact.
Related Keyword
- "Subject-adaptive meta-learning for personalized BCI: A fusion of resting-state EEG signal and task-specific information", Information Fusion, v.125
- "Efficient One-shot Federated Learning on Medical Data using Knowledge Distillation with Image Synthesis and Client Model Adaptation", Kang, Myeongkyun. (2025-10). Efficient One-shot Federated Learning on Medical Data using Knowledge Distillation with Image Synthesis and Client Model Adaptation. Medical Image Analysis, 105. doi: 10.1016/j.media.2025.103714
- "Communication Efficient Federated Learning for Multi-Organ Segmentation via Knowledge Distillation with Image Synthesis", Kim, Soopil. (2025-05). Communication Efficient Federated Learning for Multi-Organ Segmentation via Knowledge Distillation with Image Synthesis. IEEE Transactions on Medical Imaging, 44(5), 2079–2092. doi: 10.1109/TMI.2025.3525581
- "Few-shot anomaly detection using positive unlabeled learning with cycle consistency and co-occurrence features", An, Sion. (2024-12). Few-shot anomaly detection using positive unlabeled learning with cycle consistency and co-occurrence features. Expert Systems with Applications, 256. doi: 10.1016/j.eswa.2024.124890
- "Dual Attention Relation Network With Fine-Tuning for Few-Shot EEG Motor Imagery Classification", An, Sion. (2024-11). Dual Attention Relation Network With Fine-Tuning for Few-Shot EEG Motor Imagery Classification. IEEE Transactions on Neural Networks and Learning Systems, 35(11), 15479–15493. doi: 10.1109/TNNLS.2023.3287181
- "Revisiting Masked Image Modeling with Standardized Color Space for Domain Generalized Fundus Photography Classification", International Conference on Medical Image Computing and Computer Assisted Interventions, pp.538 - 548
- "MC-NuSeg: Multi-Contour Aware Nuclei Instance Segmentation with Segment Anything Model", Information Processing in Medical Imaging, IPMI 2025, pp.283 - 296
- "InstaSAM: Instance-Aware Segment Any Nuclei Model with Point Annotations", Nam, Siwoo. (2024-10-07). InstaSAM: Instance-Aware Segment Any Nuclei Model with Point Annotations. International Conference on Medical Image Computing and Computer Assisted Interventions, 232–242. doi: 10.1007/978-3-031-72083-3_22
- "Subject-Adaptive Transfer Learning Using Resting State EEG Signals for Cross-Subject EEG Motor Imagery Classification", An, Sion. (2024-10-07). Subject-Adaptive Transfer Learning Using Resting State EEG Signals for Cross-Subject EEG Motor Imagery Classification. International Conference on Medical Image Computing and Computer Assisted Interventions, 678–688. doi: 10.1007/978-3-031-72120-5_63
- "Few Shot Part Segmentation Reveals Compositional Logic for Industrial Anomaly Detection", Kim, Soopil. (2024-02-24). Few Shot Part Segmentation Reveals Compositional Logic for Industrial Anomaly Detection. AAAI Conference on Artificial Intelligence, 8591–8599. doi: 10.1609/aaai.v38i8.28703
