News
- A paper is published to IOP Machine Learning: Science and Technology
- A paper is accepted by ICML 2024
- A paper is published to Springer Applied Intelligence
- A paper is published to Springer Pattern Analysis and Applications
- A paper is published to IEEE Transactions on Artificial Intelligence
- A paper is published to Springer Applied Intelligence
- A paper is accepted by ICCV 2023
- A paper is accepted by WACV 2023
Publications
JJournal publications
CConference publications
(* indicates equal contribution)
Under review
-
CEQ-CBM: A Probabilistic Concept Bottleneck with Energy-based Models and Quantized Vectors
Sangwon Kim, Dasom Ahn, Byoung Chul Ko, In-su Jang, Kwang-Ju Kim
[paper]
-
CEnergy-Ensemble Concept Bottleneck Models: Decoding the Interpretability-Accuracy Dilemma
Dasom Ahn*, Sangwon Kim*, Hyeongjin Kim, In-su Jang, Kwang-Ju Kim, Byoung Chul Ko
[paper]
2024
-
CNeuro-Symbolic Homogeneous Concept Reasoning for Scene Interpretation
Sangwon Kim, Byoung Chul Ko, In-su Jang, Kwang-Ju Kim
Public Sector Digital Transformation Workshop, PlatCon 2024
[paper]
-
CScene Graph Generation Strategy with Co-occurrence Knowledge and Learnable Term Frequency
Hyeongjin Kim, Sangwon Kim, Dasom Ahn, Jong Taek Lee, Byoung Chul Ko
ICML 2024
[paper]
-
JConcept Graph Embedding Models for Enhanced Interpretability and Accuracy
Sangwon Kim, Byoung Chul Ko
IOP Machine Learning: Science and Technology
[paper]
-
JDomain-Free Fire Detection Using the Spatial-Temporal Attention Transform of the YOLO Backbone
Sangwon Kim, In-Su Jang, Byoung Chul Ko
Springer Pattern Analysis and Applications
[paper]
-
JBTD-RF: 3D Scene Reconstruction Using Block-Term Tensor Decomposition
Seon Bin Kim, Sangwon Kim, Dasom Ahn, Byoung Chul Ko
Springer Applied Intelligence
[paper]
2023
-
CCross-Modal Learning with 3D Deformable Attention for Action Recognition
Sangwon Kim, Dasom Ahn, Byoung Chul Ko
ICCV 2023
[paper]
-
CSTAR-Transformer: A Spatio-temporal Cross Attention Transformer for Human Action
Recognition
Dasom Ahn, Sangwon Kim, Hyunsu Hong, Byoung Chul Ko
WACV 2023
[paper]
-
JNeural Tree Decoder for Interpretation of Vision Transformers
Sangwon Kim, Byoung Chul Ko
IEEE Transactions on Artificial Intelligence
[paper]
-
JSTAR++: Rethinking Spatio-Temporal Cross Attention Transformer for Video
Action Recognition
Dasom Ahn, Sangwon Kim, Byoung Chul Ko
Springer Applied Intelligence
[paper]
2022
-
CViT-NeT: Interpretable Vision Transformers with Neural Tree Decoder
Sangwon Kim, Jaeyeal Nam, Byoung Chul Ko
ICML 2022
[paper]
-
CShift-ViT : Siamese Vision Transformer using Shifted Branches
Dasom Ahn, Hyeongjin Kim, Sangwon Kim, Byoung Chul Ko
ITC-CSCC 2022
[paper]
-
JSSL-MOT: Self-Supervised Learning Based Multi-Object Tracking
Sangwon Kim, Jimi Lee, Byoung Chul Ko
Springer Applied Intelligence
[paper]
-
JImage Registration Between Real Image and Virtual Image Based on Self-supervised
Keypoint
Learning
Sangwon Kim, In-Su Jang, Byoung Chul Ko
Springer LNCS
[paper]
-
JFacial Expression Recognition Based on Squeeze Vision Transformer
Sangwon Kim, Jaeyeal Nam, Byoung Chul Ko
MDPI Sensors
[paper]
2021
-
CImage Registration Between Real Image and Virtual Image Based on Self-Supervised
Keypoint
Learning
Sangwon Kim, In-Su Jang, Byoung Chul Ko
ACPR 2021
[paper]
-
JSelf-Supervised Keypoint Detection Based on Multi-layer Random Forest
Regressor
Sangwon Kim, Mira Jeong, Byoung Chul Ko
IEEE Access
[paper]
-
JLightweight Surrogate Random Forest Supporting for Model Simplification and
Feature Relevance
Sangwon Kim, Mira Jeong, Byoung Chul Ko
Springer Applied Intelligence
[paper]
-
JModel Simplification of Deep Random Forest for Real-Time Applications of Various
Sensor Data
Sangwon Kim, Byoung Chul Ko, Jaeyeal Nam
MDPI Sensors
[paper]
2020
-
CIs the Surrogate Model Interpretable?
Sangwon Kim, Mira Jeong, Byoung Chul Ko
Interpretable Inductive Biases Workshop, NeurIPS 2020
[paper]
-
CFast Multiple Object Tracking Using Siamese Random Forest Without Online Tracker
Updating
Jimi Lee, Sangwon Kim, Byoung Chul Ko
Benchmarking Multi-Target Tracking Workshop (MOT Challenge), CVPR 2020
[paper]
-
JBuilding Deep Random Ferns Without Backpropagation
Sangwon Kim, Byoung Chul Ko
IEEE Access
[paper]
-
JEnergy Efficient Pupil Tracking Based on Rule Distillation of Cascade Regression
Forest
Sangwon Kim, Mira Jeong, Byoung Chul Ko
MDPI Sensors
[paper]
-
JOnline Multiple Object Tracking Using Rule Distillated Siamese Random
Forest
Jimi Lee, Sangwon Kim, Byoung Chul Ko
IEEE Access
[paper]
2019
-
CDeep Coupling of Random Ferns
Sangwon Kim, Mira Jeong, Deokwoo Lee, Byoung Chul Ko
eXplainable AI Workshop, CVPR 2019
[paper]
-
CDepth-map Estimation Using Combination of Global Deep Network and Local Deep
Random Forest
SangJun Kim, Sangwon Kim, Deokwoo Lee, Byoung Chul Ko
Electronic Imaging 2019
[paper]
-
JFast Depth Estimation in a Single Image Using Lightweight Efficient Neural
Network
Sangwon Kim, Jaeyeal Nam, Byoung Chul Ko
MDPI Sensors
[paper]
Experiences
2024-now
Postdoctoral Associate, ETRI - Electronics and Telecommunications Research Institute (South Korea)
2024-2024
Postdoctoral Fellow, Keimyung University (South Korea)
2017-2018
Research and development intern, Qualcomm Institute, UC San Diego (San Diego,
CA, USA)
Winter 2017
Student intern, EPITECH - European Institute of Technology (Paris, France)
Summer 2016
Student intern, Qualcomm Institute, UC San Diego (San Diego,
CA, USA)
Education
2020-2024
Ph.D., Keimyung University (Daegu, South Korea)
Computer Engineering,
Artificial Intelligence (Advisor:
Byoung Chul Ko)
2018-2020
M.S., Keimyung University (Daegu, South Korea)
Computer Engineering, Artificial
Intelligence (Advisor:
Byoung Chul Ko)
2012-2018
B.S., Keimyung University (Daegu, South Korea)
Computer Engineering (Summa
Cum
Laude)
Academic Service
Reviewing
Journal
TNNLS, TCSVT, TMM, CVIU, TKDD, Nature Scientific Reports
Conference
CVPR, ICLR, WACV
Invited Talk
2022
Samsung Advanced Institute of Technology (SAIT)
Explainable AI: about vision and intrinsic
interpretation
Honors and Awards
2023
Best Paper Award, IEIE
2022
Best Paper Award, IEIE
2021
Research Fund, NRF, Academic Research Support Project in Science and Engineering (40M won, 2 years)
Scholarship, SAMIL Foundation (Full tuition, 3 semesters)
2020
7th Place, MOT Challenge, CVPR 2020
Best Paper Award, IEIE
Best Paper Award, KIISE
Media Covers