Augustine Cha

E-mail : augustine.h.cha@gmail.com LinkedIn : https://www.linkedin.com/in/augustinecha/

Updated: Sept. 21, 2021

Interests

• Depth image enhancement, Time-of-Flight Sensors, Computer Vision, Machine Learning

Education

   Columbia University
   M.S. in Computer Science
   Aug. 2019 ~ Jan. 2021
   Research Assistant

   Konkuk University
   B.S. in Electronics Engineering
   Mar. 2011 ~ Feb. 2019
   Undergraduate Researcher

Skills

• Languages: C++, Python, MATLAB, LaTeX, Markdown
• Frameworks: OpenCV, PyTorch, TensorFlow
• Others: Scuba diving(PADI Adv. Open Water), Free diving(SSI Level 3)

Work Experience

• Research and Development Engineer Jan. 2021 ~ Present
  Microsoft, Mountain View, California

• R&D Engineering Intern June. 2020 ~ Aug. 2020
  Microsoft, Mountain View, California
  • Worked as a research and development engineering intern at Microsoft Time-of-Flight Research Team.
  • Developed a robust and power-efficient system for detection and localization of the sensor and multiple objects with passive and active imaging.
  • The framework developed to be flexible of using different sensors(e.g., IMU) and systems with low memory.
  • Developed a statistical approach to detect moving objects and track the objects using EKF.
  • Capable under both static or dynamic scenes using optical flow detection using feature matching
  • Capable of calculating the 6 degrees of freedom of the sensor and the moving object.
• Software Engineer Dec. 2018 ~ Jun. 2019
  Analogue plus, Seongnam, South Korea
  • Participating on a Smart Farm project to develop an autonomous harvesting system. Applying computer vision and deep learning algorithms to detect fruits in orchards.
  • Developing path finding algorithms for a robot-arm to harvest detected fruits.
  • Creating a GUI program for object detection using computer vision and deep learning algorithms.

Research Experience

• Graduate Research Assistant Feb. 2020 ~ Dec. 2020
  Columbia University
  • Currently working on invariant/equi-variant object detection for computer vision using meta learning techniques.
• Undergraduate Researcher Nov. 2016 ~ Jun. 2018
  Konkuk University
  • Applied saliency detection as a filtering weight to guide feature matching algorithms to increase matching reliability. Published an international paper and gave a poster presentation based on the work.
  • Collaborated with a colleague in a project on haze removal. Applied multi-scale superpixel using SLIC to establish a precise Dark Channel Prior(DCP) map. Published a domestic paper and gave a poster presentation based on the work.
  • Conducted a self-project on a learning-based feature point detector by mimicking handcraft feature detecting algorithms using neural network. Applied auto-encoding networks to learn feature points produced by SIFT.

Publications

• Generative Interventions for Causal Learning
  Chengzhi Mao, Augustine CHA, Amogh Gupta, Hao Wang, Junfeng Yang, and Carl Vondrick
  IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR), Aug. 2020 [PDF]

• Saliency-guided feature matching for self-driving systems
  Augustine H. CHA and Wonjun Kim
  IEEE International Conference on Consumer Electronics-Asia(ICCE-Asia), Jun. 2018 [PDF]