0%

About Me

Chang Huai-Yuan

Email: hamiltonchangwork”at”gmail.com / hamiltonchang”at”pku.edu.cn

Education

  • Peking University, M.E. in Software Engineering, Sept. 2017 - Aug. 2020

  • Chung Yuan Christian University, B.S. in Computer Science and Information Engineering, Sept. 2013 - June 2017

Experience

  • Infortrend Technology, New-Taipei, Taiwan

    AI Research and Develop Engineer, May.2021-Jul.2022

    • Solved the issue of multi-processes shared face database through shared memory and accelerated comparison time by 3times, and accelerated inference time by 5 times through quantization and TensorRT on Face Recognition API
    • Reduced half docker image size of AI Service API through Multi-Stage and deleted-redundant dependencies.
    • Reduced inference time by 3 times on general CPU platforms using TensorFlow Lite with XNNPack.
    • Developed an auto-tiered model to improve IO performance by 60% on Infortrend storage systems
  • Patere Technologies, Inc., Taipei, Taiwan

    Computer Vision Engineer, Nov. 2020 - Jan. 2021

    • Designed a rule-based confidence and speech fluency detection algorithm with basic speech features
    • Built an extractor to get more complex and useful features which served for ML/DL, such as MFCC, FBank features
    • Designed an attention detection algorithm reaching about 80% accuracy, precision and recall on the testing dataset
  • Lenovo Group Ltd, Lenovo Research, Beijing, China

    Computer Vision and AI Algorithm Intern, Aug. 2019 - Nov. 2019

    • Developed automatic training system for commodity detection based on RetinaNet for unmanned stores
    • Optimzed the success rate of commodity detection about 20% through collecting and generating more complex and suitable data
  • Zero Zero Robotics, Beijing, China

    Computer Vision and AI Algorithm Intern, Nov. 2018 - Aug. 2019

    • Developed, maintained and optimized long-term object tracking module on Hover Camera 2 drone
    • Improved by 15% tracking success rate and reduced by 5% CPU utility through adding constraints and modifying object re-identification strategy for tracking module

Project

  • Design and Implementation of Real-time Long-term Single Person Tracking System

    Individual Work, Nov. 2019 - April. 2020

    • Tracking successful rate is 79.02%, recall is 77.54%, and FPS is 12.87 on Intel i7-6700k CPU
    • Built three tracking states: target selection, track and retrieval, to fully handle the situations
    • Designed several strategies to achieve real-time and stable long-term person tracking
  • 2D Anime Charactors Recognition

    Team Leader, May 2018 - June 2018

    • TOP1 result is 73.25%
    • Developed a model to recognize the different anime characters and let it can distinguish the same ones drawn by distinct artists
    • Utilized the trained LBPcascade to detect anime faces and recognized ones with the fine-tuned Inception-v3 model
  • OurScheme: A Scheme Interpreter Implemented in C++

    Individual Work, Mar. 2017 - May 2017

    • Developed a complete interpreter workflow
    • Designed for high stability and robustness
  • Face Recognition Access Control System Based on Raspberry Pi

    Team Leader,  Feb. 2016 - Nov. 2016

    • Developed a complete face recognition access control solution on Raspberry Pi
    • Recall and precision is about 85%

Skills

  • Programming Language: Python, C/C++, Java, R
  • Other: Git, Docker, Kubernetes, GDB, pdb