Vyz1k39itirzpqm9ade4

莊承翰

The devil is in the detail.

目前專研在deep learning,為團隊尋找/驗證/分析各種新方法,並結合不同方法到實際的場域中,銜接學術理論到實際應用。

Algorithm Engineer
城市,TW
[email protected]

Summary


  • Summary 5 years+ experience of Deep Learning, 9 years experience of Computer vision / Image processing. 
  • Capable of building AI(Image/Video) product from scratch both individually or co-work with various teams. 
  • Analysis, Experiment, Debug, Improve the Deep Learning system.

Kaggle


  • Top 1% ( 16 / 1559 ) Sartorius - Cell Instance Segmentation (2021/12~2021/12)
  • Top 4% ( 85 / 2160 ) Human Protein Atlas Image Classification (2018/11~2019/1)
    profile

Skills


Software

  • C++
  • Python


Library / Framework

  • OpenCV
  • PyTorch
  • LibTorch


DL / ML

  • Classification / Regression
  • One stage / Two stage Object Detection 
  • Inbalance Data Handle 
  • Domain Adaptation 
  • Semi/Self Supervised 
  • Data Augmentation 
  • Mislabel Data Detection 
  • New Class Detection
  • Abnormal Detection

Experience

Delta Electronics, Algorithm Engineer, Jul 2017 ~ Now

[Projects]

  • TFT-LCD defect detection
  • Capacitance defect detection
  • Ceramic substrate defect detection 
  • POC projects x 6
  • Algorithms system Development & Maintain

[Achievement]

  • Improve the accuracy of algorithms (compared with the previous version)
      [TFT-LCD]
           Achieve the accuracy 85-95%  (human baseline~= 88~95% depend on different product line)
      [Capacitance]
           Improved the accuracy from 76% to 92% (human baseline = 94%) 
      [Ceramic substrate]
           Achieve the accuracy 93%    (human baseline = 94%)
  • Deal with the problem of the same product on the different machines, make the model can deploy to multiple machines
  • Conversion metric make the solution more able to meet customer needs
  • Provide the innovative methods for team members to help other projects proceed smoothly

 [Skills]

  • Deep Learning (Pytorch)
  • Transitional CV
  • C++ / Python

ADE Technology, Algorithm Engineer, Apr 2016 ~ Jul 2017 

[Projects]

  • License Plate Recognition System
    • Algorithms development
    • UI
  • Pedestrian Detection
    • Algorithms development
  • Thermal Image Enhancement
 [Skills] 

  • Transitional CV 
  • C++ / Python

Genius Vision Digital, Algorithm Engineer, Oct 2014 ~ Feb 2016

[Projects]

  • License Plate Algorithms 
  • People Counting Algorithms 
  • System Debug and Maintain
[Achievement]

  • Significantly improve the executive performance of people counting algorithms (Reduce the CPU loading by 50%)
 [Skills] 

  • Transitional CV 
  • C++

Education

National Chin-Yi University of Technology, 2011 ~ 2013

Master of Computer Science and Information Engineering

National Chin-Yi University of Technology, 2007 ~ 2011

Bachelor of  Computer Science and Information Engineering

Powered By CakeResume