Publications

For recent research papers, please see Google Scholar.

Note: I'm no longer maintaining this page. Please find recent update here: ailab.khu.ac.kr/pub/

*: Primary author (first/corresponding author)

[J]: Journal, [C]: Conference, [T]: Technical Report (arXiv preprint), [P]: Registered Patent, [PA]: Patent Application

2020

[J11] CUA Loss: Class Uncertainty-Aware Gradient Modulation for Robust Object Detection

• • Jung Uk Kim, Seong Tae Kim, Hong Joo Lee, Sangmin Lee, and Yong Man Ro

  • IEEE Transactions on Circuits and Systems for Video Technology (TCSVT)

[J10] Force-Ultrasound Fusion: Bringing Spine Robotic-US to the Next "Level"

• • Maria Tirindelli, Maria Victorova, Javier Esteban, Seong Tae Kim, David Navarro-Alarcon, Yong Ping Zheng, Nassir Navab

  • IEEE Robotics and Automation Letters (RA-L)

  • Presented at IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

[J9] Multimodal Facial Biometrics Recognition: Dual-stream Convolutional Neural Networks with Multi-feature Fusion Layers

• • Leslie Ching Ow Tiong, Seong Tae Kim, and Yong Man Ro

  • Image and Vision Computing (ImaVis)

[J8] Lightweight and effective facial landmark detection using adversarial learning with face geometric map generative network

• • Hong Joo Lee, Seong Tae Kim, Hakmin Lee, and Yong Man Ro

  • IEEE Transactions on Circuits and Systems for Video Technology (TCSVT)

[C27] Self-supervised out-of-distribution detection in brain CT scans

• • Abinav Ravi Venkatakrishnan, Seong Tae Kim*(co-first), Rami Eisawy, Franz Pfister, Nassir Navab

  • Medical Imaging Meets NeurIPS (NeurIPS Workshop)

[C26] TeCNO: Surgical Phase Recognition with Multi-Stage Temporal Convolutional Networks

• • Tobias Czempiel, Magdalini Paschali, Matthias Keicher, Walter Simson, Hubertus Feussner, Seong Tae Kim, Nassir Navab

  • International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI) [code]

[C25] Spatio-temporal Learning from Longitudinal Data for Multiple Sclerosis Lesion Segmentation

• • Stefan Denner, Ashkan Khakzar, Moiz Sajid, Mahdi Saleh, Ziga Spiclin, Seong Tae Kim*, Nassir Navab

  • BrainLes at International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAIW) [code]

[C24] Robust Ensemble Model Training via Random Layer Sampling Against Adversarial Attack

• • Hakmin Lee, Hong Joo Lee, Seong Tae Kim, and Yong Man Ro

  • British Machine Vision Conference (BMVC)

[C23] Towards high-performance object detection: Task-specific design considering classification and localization separation

• • Jung Uk Kim, Seong Tae Kim, Eun Sung Kim, Sang-Keun Moon, and Yong Man Ro

  • International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

[T5] Efficient Ensemble Model Generation for Uncertainty Estimation with Bayesian Approximation in Segmentation

• • Hong Joo Lee, Seong Tae Kim*(co-first), Hakmin Lee, Nassir Navab, Yong Man Ro

  • arxiv Preprint arXiv:2005.10754

[T4] Confident Coreset for Active Learning in Medical Image Analysis

• • Seong Tae Kim*, Farrukh Mushtaq(co-first), Nassir Navab

  • arxiv Preprint arXiv:2004.02200

[P10] Automated Facial Expression Recognizing Systems on N frames, Methods, and Computer-Readable Mediums thereof

• KR 10-2152120

[P9] Machine learning system using joint learning and the method thereof

• KR 10-2100973

[P8] Method for generating breast masses according to the lesion characteristics and the system thereof

• KR 10-2067340

2019

[J7] Attended relation feature representation of facial dynamics for facial authentication

• • Seong Tae Kim* and Yong Man Ro

  • IEEE Transactions on Information Forensics and Security (TIFS)

[J6] Implementation of multimodal biometric recognition via multi-feature deep learning networks and feature fusion

• • Leslie Ching Ow Tiong, Seong Tae Kim, and Yong Man Ro

  • Multimedia Tools and Applications (MTAP)

[C22] Building a breast-sentence dataset: Its usefulness for computer-aided diagnosis

• • Hyebin Lee, Seong Tae Kim, and Yong Man Ro

  • Visual Recognition for Medical Image at International Conference on Computer Vision (ICCVW), South Korea

[C21] Realistic breast mass generation through BIRADS category

• • Hakmin Lee, Seong Tae Kim, Jae-Hyeok Lee, and Yong Man Ro

  • International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), China

[C20] Generation of multimodal generation using visual word constraint model for explainable computer-aided diagnosis

• • Hyebin Lee, Seong Tae Kim, and Yong Man Ro

  • Workshop on Interpretability of Machine Intelligence in Medical Image Computing at MICCAI (MICCAIW), China [Demo]

[C19] Probenet: Probing deep networks

• • Jae-Hyeok Lee, Seong Tae Kim, and Yong Man Ro

  • IEEE International Conference on Image Processing (ICIP), Taiwan

[C18] Visual evidence for interpreting diagnostic decision of deep neural network in computer-aided diagnosis

• • Seong Tae Kim*, Jae-Hyeok Lee, and Yong Man Ro

  • SPIE Medical Imaging (SPIE MI), USA

[P7] Analysis method of relations of face movements and the system thereof

• KR 10-2054058

[P6] Method and system for artificial intelligence based video surveillance using deep learning

• KR 10-1995107

[P5] Explainable computer-aided diagnosis and the method thereof

• KR 10-1938992

2018

[C17] Facial dynamics interpreter network: What are the important relations between local dynamics for facial trait estimation?

• • Seong Tae Kim* and Yong Man Ro

  • European Conference on Computer Vision (ECCV), Germany [poster][Demo]

[J5] Visually interpretable deep network for diagnosis of breast masses on mammograms

• • Seong Tae Kim*, Jae-Hyeok Lee, Hakmin Lee, and Yong Man Ro

  • Physics in Medicine and Biology (PMB)

[C16] Feature2Mass: Visual feature processing in latent space for realistic labeled mass generation

• • Jae-Hyeok Lee, Seong Tae Kim, Hakmin Lee, and Yong Man Ro

  • Bioimage Computing Workshop at European Conference on Computer Vision (ECCVW), Germany [poster]

[C15] ICADx: Interpretable computer aided diagnosis of breast masses

• • Seong Tae Kim*, Hakmin Lee, Hak Gu Kim, and Yong Man Ro

  • SPIE Medical Imaging (SPIE MI), USA (Robert F. Wagner All Conference Best Student Paper Final Lists Award, Oral presentation)

[C14] Convolution with logarithmic filter groups for efficient shallow CNN

• • Tae Kwan Lee, Wissam J Baddar, Seong Tae Kim, Yong Man Ro

  • International Conference on Multimedia Modeling (MMM), Thailand (Oral presentation)

[C13] Teacher and student joint learning for compact facial landmark detection network

• • Hongju Lee, Wissam Baddar, Hak Gu Kim, Seong Tae Kim, and Yong Man Ro

  • International Conference on Multimedia Modeling (MMM), Thailand (Oral presentation)

[P4] Apparatus and method for generating reprojection images for diagnostic feature extraction

• US 10,092,263

[P3] System for instructional video learning and evaluation using deep learning

• KR 10-1893290

2017

[J4] Latent feature representation with depth directional long-term recurrent learning for breast masses in digital breast tomosynthesis

• • Dae Hoe Kim, Seong Tae Kim, Jung Min Chang, and Yong Man Ro

  • Physics in Medicine and Biology (PMB)

[C12] Multi-scale facial scanning via spatial LSTM for latent facial feature representation

• • Seong Tae Kim*, Yeoreum Choi, and Yong Man Ro

  • International Conference of the Biometrics Special Interest Group (BIOSIG) , Germany

[C11] Adaptive attention fusion network for visual question answering

• • Geonmo Gu, Seong Tae Kim, and Yong Man Ro

  • IEEE International Conference on Multimedia and Expo (ICME), Hong Kong [Demo]

[T2] Differential generative adversarial networks: Synthesizing non-linear facial variations with limited number of training data

• • Geonmo Gu, Seong Tae Kim, Kihyun Kim, Wissam J Baddar, Yong Man Ro

  • arxiv Preprint arXiv:1711.10267

[T1] EvaluationNet: Can human skill be evaluated by deep networks?

• • Seong Tae Kim* and Yong Man Ro

  • arxiv Preprint arXiv:1705.11077

[P2] Method and system for automatic biometric authentication based on facial spatio-temporal features

• KR 10-1802061

[P1] Lesion classification apparatus, and method of modifying lesion classification data

• US 9,547,896

2016

[C10] Facial dynamic modelling using long short-term memory network: Analysis and application to face authentication

• • Seong Tae Kim*, Dae Hoe Kim, and Yong Man Ro

  • IEEE International Conference on Biometrics: Thoery, Applications, and Systems (BTAS), USA [poster] [Demo]

[C9] Spatio-temporal representation for face authentication by using multi-task learning with human attributes

• • Seong Tae Kim*, Dae Hoe Kim, and Yong Man Ro

  • IEEE International Conference on Image Processing (ICIP), USA [poster]

[C8] A deep facial landmarks detection with facial contour and facial components constraint

• • Wissam Baddar, Jisoo Son, Dae Hoe Kim, Seong Tae Kim, Yong Man Ro

  • IEEE International Conference on Image Processing (ICIP), USA [poster]

[C7] Latent feature representation with 3-D Multi-view convolutional neural network for bilateral analysis in digital breast tomosynthesis

• • Dae Hoe Kim, Seong Tae Kim, and Yong Man Ro

  • IEEE International Conference on Acoustics, speech and signal processing (ICASSP), China

2015

[J3] Detection of masses in digital breast tomosynthesis using complementary information of simulated projection

• • Seong Tae Kim*, Dae Hoe Kim, and Yong Man Ro

  • Medical Physics (MP)

[J2] Improving mass detection using combined feature representations from projection views and reconstructed volume of DBT and boosting based classification with feature selection

• • Dae Hoe Kim, Seong Tae Kim, and Yong Man Ro

  • Physics in Medicine and Biology (PMB) [Demo]

[C6] Region matching based on local structure information in ipsilateral digital breast tomosynthesis views

• • Seong Tae Kim*, Dae Hoe Kim, Dong Jin Ji, and Yong Man Ro

  • IEEE International Conference on Image Processing (ICIP), Canada [poster]

[C5] Feature extraction from bilateral dissimilarity in digital breast tomosynthesis reconstructed volume

• • Dae Hoe Kim, Seong Tae Kim, Wissam J Baddar, and Yong Man Ro

  • IEEE International Conference on Image Processing (ICIP), Canada (Selected as Top 10% paper) [poster]

[C4] Combination of conspicuity improved synthetic mammograms and digital breast tomosynthesis: A promising approach for mass detection

• • Seong Tae Kim*, Dae Hoe Kim, and Yong Man Ro

  • SPIE Medical Imaging (SPIE MI), USA (Oral presentation)

[C3] Feature extraction from inter-view similarity of DBT projection views

• • Dae Hoe Kim, Seong Tae Kim, and Yong Man Ro

  • SPIE Medical Imaging (SPIE MI), USA (Best poster award) [poster]

2014

[J1] Breast mass detection using slice conspicuity in 3D reconstructed digital breast volumes

• • Seong Tae Kim*, Dae Hoe Kim, and Yong Man Ro

  • Physics in Medicine and Biology (PMB), 2014

[C2] Generation of conspicuity-improved synthetic image from digital breast tomosynthesis

• • Seong Tae Kim*, Dae Hoe Kim, and Yong Man Ro

  • International Conference on Digital Signal Processing (DSP), Hong Kong, 2014 (Oral presentation)

[C1] Mass detection based on pooled mass probability map of 3D reconstructed slices in digital breast tomosynthesis

• • Seong Tae Kim*, Dae Hoe Kim, Eun Suk Cha, and Yong Man Ro

  • IEEE-EMBS International Conference on Biomedical and Health Informatics (BHI), Spain, 2014 (Oral presentation)