Soonmin Hwang
Ph.D. Candidate
Department of Electrical Engineering

Email: smhwang [at] rcv.kaist.ac.kr, jjang9hsm [at] gmail.com


Research Interests

  • Fast-/Real-world object detection algorithms
  • Deep learning: adversarial training, domain adaptationsed learning of deep network
  • 3D scene understanding
  • Multi-sensor system
    • Visible + Longwave-length IR (Thermal image)
    • 3D- LiDAR (Velodyne)

Publications    

2016 
 A New Perspective on Thermal Imaging Enhancement using Convolutional Neural Network
Yukyung ChoiNamil KimSoonmin Hwang and In So Kweon
Patent Pending* in Korea (10-2016-0100058), Winner in KINPEC 2016.
[website]

 
KAIST Multi-spectral Recognition Dataset in day and night
The International Journal of Robotics Research (IJRR), 2016. (under review).
[pdf] [project page] [video]
Thermal Image Enhancement using Convolution Neural Network
Soonmin Hwang*Yukyung Choi*Namil Kim*and In So Kweon
The International Conference of Intelligent Robots and Systems (IROS), Oct 2016.
1st place in NVidia Deep Learning Contest 2016
[pdf] [project page]
 
 
Fast Multiple Objects Detection and Tracking Fusing Color Camera and 3D LIDAR for Intelligent Vehicles
Soonmin Hwang*Namil Kim*Yukyung ChoiSeokju Lee, and In So Kweon
International Conference on Ubiquitous Robots and Ambient Intelligence (URAI), Aug 2016. 
*equal contributions
[pdf] [project page] [bibtex]
  
 Thermal-Infrared based Drivable Region Detection
Jae Shin Yoon, Kibaek Park, Namil Kim, Soonmin Hwang, Yukyung Choi, Francois Rameau, and In So Kweon
Intelligent Vehicles Symposium (IV), Jun 2016. 
[pdf] 
  
2015
Low-Cost Synchronization for Multispectral Cameras
Soonmin HwangYukyung ChoiNamil Kim, Kibaek Park, Jae Shin Yoon, and In So Kweon
International Conference on Ubiquitous Robots and Ambient Intelligence (URAI), Nov 2015.
[pdf] [bibtex]
  
Geometrical Calibration of Multispectral Calibration
Namil KimYukyung ChoiSoonmin Hwang, Kibaek Park, Jae Shin Yoon, and In So Kweon
International Conference on Ubiquitous Robots and Ambient Intelligence (URAI), Nov 2015.
[pdf] [bibtex]
  
ARTRIEVAL: Painting Retrieval Without Expert Knowledge
Namil KimYukyung ChoiSoonmin Hwang, and In So Kweon
IEEE International Conference on Image Processing (ICIP), Sep 2015.
[pdf] [bibtex] [project page]
  
All-Day Visual Place Recognition: Benchmark Dataset and Baselines
Yukyung ChoiNamil Kim, Kibaek Park, Soonmin Hwang, Jae Shin Yoon, and In So Kweon
IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPR-VPRICE), Jun 2015.
[pdf] [bibtex] [project page]
  
Multispectral Pedestrian Detection: Benchmark Dataset and Baselines
Soonmin HwangJaesik ParkNamil KimYukyung Choi, and In So Kweon
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Jun 2015.
[pdf] [bibtex] [project page]


2014
A Two Phase Approach for Pedestrian Detection
Soonmin hwang, Tae-hyun Oh, and In So Kweon
Asian Conference on Computer Vision Workshops (ACCV-IVVT), Nov 2014.
[pdf] [bibtex]


2013

Evaluation of Vocabulary Trees for Localization in Robot Applications
Soonmin Hwang, Chaehoon Park, Yukyung Choi, Donggeun Yoo, and In So Kweon
In Proc. of International Conference on Control, Automation and Systems (ICCAS), Oct 2013.
[pdf]
























































shopify visitor statistics