|
Taejune Kim Research Engineer at Robotics Lab, Hyundai Motor Company |
|
ResearchI'm interested in computer vision, anomaly detection, and object detection. Representative papers are highlighted. |
|
Continuous Memory Representation for Anomaly Detection
Joo Chan Lee*, Taejune Kim*, Eunbyung Park, Simon S. Woo, Jong Hwan Ko ECCV, 2024 project page / code Unified framework for unsupervised anomaly detection. |
|
Patch-wise Vector Quantization for Unsupervised Medical Anomaly Detection
Taejune Kim, Yun-Gyoo Lee, Inho Jeong, Soo-Youn Ham, Simon S. Woo Pattern Recognition Letters, 2024 (Accepted) We propose a patch-wise vector quantization to perform high-accuracy unsupervised anomaly detection in CT images. |
|
Rotated-DETR: an End-to-End Transformer-based Oriented Object Detector for Aerial Images
Jinbeom Kim*, Giljun Lee*, Taejune Kim, Simon S. Woo SAC, 2023 Inferencing oriented bounding box using Deformable-DETR framework. |
|
MGCMA: Multi-scale Generator with Channel-wise Mask Attention to generate Synthetic Contrast-enhanced Chest Computed Tomography
Jeongho Kim, Yun-Gyoo Lee, Donggeun Ko, Taejune Kim, Soo-Youn Ham, Simon S. Woo SAC, 2023 Synthesizing contrast-enhanced style on non-contrast CT scans. |
|
A2: Adaptive Augmentation for Effectively Mitigating Dataset Bias
Jaeju An, Taejune Kim, Donggeun Ko, Sangyup Lee, Simon S. Woo ACCV, 2022 Mitigating dataset bias through domain adaptation. |
|
Evading Deepfake Detectors via High Quality Face Pre-Processing Methods
Jeongho Kim*, Taejune Kim*, Jeonghyeon Kim, Simon S. Woo ICPR, 2022 Concealing deepfake artifacts via image processing procedures. |
Awards
2nd Place, Artificial Intelligence Grand Challenge 2022 (AGC 2022) - 4th Competition Track 3
|
|
Design and source code from Jon Barron's website. |