Myungjoo Kang

ORCID: 0000-0002-8064-7167
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Image and Signal Denoising Methods
  • Advanced Image Processing Techniques
  • Sparse and Compressive Sensing Techniques
  • Computational Fluid Dynamics and Aerodynamics
  • Computer Graphics and Visualization Techniques
  • Medical Image Segmentation Techniques
  • Advanced Neural Network Applications
  • Fluid Dynamics and Turbulent Flows
  • Advanced Numerical Methods in Computational Mathematics
  • Advanced Numerical Analysis Techniques
  • Generative Adversarial Networks and Image Synthesis
  • Advanced Image Fusion Techniques
  • Lattice Boltzmann Simulation Studies
  • Model Reduction and Neural Networks
  • 3D Shape Modeling and Analysis
  • Stochastic processes and financial applications
  • Fluid Dynamics and Heat Transfer
  • Neural Networks and Applications
  • Anomaly Detection Techniques and Applications
  • Video Surveillance and Tracking Methods
  • Adversarial Robustness in Machine Learning
  • Domain Adaptation and Few-Shot Learning
  • Numerical methods in inverse problems
  • Image Processing Techniques and Applications
  • Stock Market Forecasting Methods

Seoul National University
2015-2024

National Institute for Mathematical Sciences
2015-2024

Mathematical Sciences Research Institute
2006-2024

University of Seoul
2023

Seoul Social Welfare Graduate University
2021

York University
2020

TU Wien
2007-2008

Level Set Systems (United States)
2002-2005

University of California, Los Angeles
1999-2002

10.1023/a:1011178417620 article EN Journal of Scientific Computing 2000-01-01

Multi-dye-sensitized upconverting nanoparticles (UCNPs), which harvest photons of wide wavelength range (450–975 nm) are designed and synthesized. The UCNPs embedded in a photo-acid generating layer integrated on destructible nonvolatile resistive memory device. Upon illumination light, the system permanently erases stored data, achieving enhanced information security. As service to our authors readers, this journal provides supporting supplied by authors. Such materials peer reviewed may be...

10.1002/adma.201603169 article EN Advanced Materials 2016-10-17

Image processing via convolutional neural network (CNN) has been developed rapidly for remote sensing technology. Moreover, techniques accurately extracting building footprints from sensed images have attracted considerable interest owing to their wide variety of common applications, including monitoring natural disasters and urban development. Extraction can be performed easily by semantic segmentation using U-Net-like CNN architectures. However, obtaining precise boundaries masks remains...

10.1109/tgrs.2021.3108781 article EN IEEE Transactions on Geoscience and Remote Sensing 2021-09-02

In this paper, we propose a new capsule network architecture called Attention Routing CapsuleNet (AR CapsNet). We replace the dynamic routing and squash activation function of with (CapsuleNet) attention activation. The is between capsules through an module. fast forward-pass while keeping spatial information. On other hand, intuitive interpretation finding centroid prediction capsules. Thus, its variant focus on preserving vector orientation. However, focuses performing capsule-scale...

10.1109/iccvw.2019.00247 preprint EN 2019-10-01

This paper reviews the NTIRE 2020 challenge on real image denoising with focus newly introduced dataset, proposed methods and their results. The is a new version of previous 2019 that was based SIDD benchmark. collected validation testing datasets, hence, named SIDD+. has two tracks for quantitatively evaluating performance in (1) Bayer-pattern rawRGB (2) standard RGB (sRGB) color spaces. Each track ~250 registered participants. A total 22 teams, proposing 24 methods, competed final phase...

10.1109/cvprw50498.2020.00256 article EN 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) 2020-06-01

10.1016/j.jvcir.2013.01.010 article EN Journal of Visual Communication and Image Representation 2013-01-19

A level-set model for the simulation of epitaxial growth is described. In this model, motion island boundaries discrete atomic layers determined by time evolution a continuous function \ensuremath{\varphi}. The adatom concentration treated in mean-field manner. We use to systematically examine importance various fluctuations submonolayer and multilayer regimes. find that, regime large values $D/F,$ dominant are associated with spatial seeding islands. also show how different microscopic...

10.1103/physrevb.65.195403 article EN Physical review. B, Condensed matter 2002-04-19

10.1016/j.camwa.2015.12.038 article EN Computers & Mathematics with Applications 2016-01-01

Purpose.: To introduce a novel, digital, three-dimensional (3D) reconstruction of the optic nerve head (ONH) and to use this method evaluate 3D configuration lamina cribrosa (LC) in patients with primary open-angle glaucoma. Methods.: Optic discs 137 eyes glaucoma were scanned enhanced depth-imaging spectral domain-optical coherence tomography (SD-OCT). images ONH then reconstructed from B-scan using maximum intensity projection (MIP) texture-based volume rendering (VRT). The performance...

10.1167/iovs.11-7848 article EN Investigative Ophthalmology & Visual Science 2011-12-14

With more than 40% of the world’s population at risk, 200–300 million infections each year, and an estimated 1.2 deaths annually, malaria remains one most important public health problems mankind today. propensity parasites to rapidly develop resistance newly developed therapies, recent failures artemisinin-based drugs in Southeast Asia, there is urgent need for new antimalarial compounds with novel mechanisms action be against multidrug resistant malaria. We present here a image analysis...

10.1371/journal.pone.0061812 article EN cc-by PLoS ONE 2013-04-23

We present a customized high content (image-based) and throughput screening algorithm for the quantification of Trypanosoma cruzi infection in host cells. Based solely on DNA staining single-channel images, precisely segments identifies nuclei cytoplasm mammalian cells as well intracellular parasites infecting The outputs statistical parameters including total number cells, infected per image, average cell, ratio (defined divided by cells). Accurate precise estimation these allow both...

10.1371/journal.pone.0087188 article EN cc-by PLoS ONE 2014-02-04

10.1016/j.cam.2016.09.024 article EN publisher-specific-oa Journal of Computational and Applied Mathematics 2016-10-01

Mechanical defects in real situations affect observation values and cause abnormalities multivariate time series, such as sensor or network data. To perceive data, it is crucial to understand the temporal context interrelation between variables simultaneously. The anomaly detection task for especially unlabeled has been a challenging problem, we address by applying suitable data degradation scheme self-supervised model training. We define four types of synthetic outliers propose which...

10.48550/arxiv.2305.04468 preprint EN other-oa arXiv (Cornell University) 2023-01-01

10.1016/j.cviu.2023.103720 article EN Computer Vision and Image Understanding 2023-05-06

We introduce Multimodal Matching based on Valence and Arousal (MMVA), a tri-modal encoder framework designed to capture emotional content across images, music, musical captions. To support this framework, we expand the Image-Music-Emotion-Matching-Net (IMEMNet) dataset, creating IMEMNet-C which includes 24,756 images 25,944 music clips with corresponding employ multimodal matching scores continuous valence (emotional positivity) arousal intensity) values. This score allows for random...

10.48550/arxiv.2501.01094 preprint EN arXiv (Cornell University) 2025-01-02

In this paper, we propose the neural shortest path (NSP), a vector-valued implicit representation (INR) that approximates distance function and its gradient. The key feature of NSP is to learn exact (ESP), which directs an arbitrary point nearest on target surface. decomposed into magnitude direction, variable splitting method used each component gradient, respectively. Unlike existing methods learning itself, ensures simultaneous recovery We mathematically prove guarantees convergence in...

10.48550/arxiv.2502.06047 preprint EN arXiv (Cornell University) 2025-02-09
Coming Soon ...