Profile

Jiun Bae배지운

Driven machine learning enthusiast specializing in natural language processing and computer vision, committed to solving real-world problems through data-driven innovation.

Experience

Return ZeroJul. 2024 - Present

Research Engineer

  • Pioneered research and development of SOTA ASR models by leveraging extensive user data. Designed and implemented a robust, scalable production pipeline.

NCSOFTGraphics AI Lab.Feb. 2021 - Jun. 2024

Machine Learning Researcher

  • Spearheaded the implementation of preprocessing and alignment pipelines for high-fidelity motion and facial capture data, significantly enhancing data quality.
  • Pioneered the development of a PyTorch-based framework for deep learning model training and experiment management, substantially boosting team productivity and streamlining research processes.
  • Led the research on a deep learning model for generating speech-based 3D facial animation, contributing to advancements in digital human technology.

NaverClovaJul. 2020 - Aug. 2020

Machine Learning Engineer, Intern

  • Contributed to the development of a sophisticated depth estimation network for 3D map reconstruction.

WebtoonResearchSep. 2018 - Feb. 2019

Machine Learning Researcher, Intern

  • Developed an advanced annotation tool for 2D bounding box support, integrating features such as live data model inference, dataset management with custom query types, and various tagging options.
  • Designed the Visual QA task for Workshop, developing a multi-modal network and dataset for its validation.

SW Maestro7thJul. 2016 - Aug. 2018

  • Garnered acclaim as a top team, showcased at the "100+ Conference" by the Ministry of Science and ICT, highlighting exceptional project development and execution.

Projects

Text-to-Texture Project
NCSOFTDec. 2023 - Jun. 2024

  • Developed a scalable infrastructure, including model serving capabilities.
  • Spearheaded the development of a Proof-of-Concept (PoC) web application using React.js and three.js, along with a FastAPI server.
  • Conducted extensive research and experimentation on models for generating natural textures on 3D objects.

Digital Human Project
NCSOFTFeb. 2022 - Nov. 2023

  • Led the development of a real-time and responsive facial expression generation model and system for digital humans.
  • Pioneered the generation of dynamic facial expressions and persona integration in conversational scenarios, enhancing digital human interactivity.
  • Orchestrated the creation of highly detailed facial animations, capturing nuanced expressions tailored to unique character traits.
  • Architected a high-performance API server, optimizing it for diverse inputs in facial animation generation, ensuring scalability and robustness.

3D Facial Animation Generation Model Research
NCSOFTFeb. 2021 - Jan. 2022

  • Implemented advanced preprocessing and alignment pipelines for superior motion and facial capture data quality.
  • Devised and deployed a PyTorch-based framework for deep learning model training, elevating team productivity and streamlining research management.
  • Conducted innovative research on deep learning models for speech-driven 3D facial animation generation.

Webtoon Annotation Tool
Naver WebtoonSep. 2018 - Feb. 2019

  • Engineered advanced annotation tools for efficient collection of domain-specific raw data.
  • Developed a sophisticated analysis tool for cross-validating annotations, enabling comprehensive tracking of productivity, task difficulty assessment, and model inference comparison.

MariaDB Scalable Lock Manager
ITE4065Dec. 2017

  • Enhanced the InnoDB storage engine's performance by replacing a lock-free design and reusable object pool for the global lock mechanism in the transaction record manager.

Memento: All Information About That Person
SW MaestroApr. 2017 - Oct. 2017

  • Created a robust, scalable crawling system to aggregate a wide array of articles and information.
  • Designed a sophisticated system for vector embedding and clustering, linking information to entities and events, while allowing for continuous vector expansion.

DeepCheck: Deep Learning-Based Auto Face Recognition Attendance Check
SW MaestroSep. 2016 - Dec. 2016

  • Enhanced a face recognition model specifically for low-resolution, crowded environments.
  • Developed a cross-platform application and API server using React Native, enabling efficient, lightweight model inference.

Education

Hanyang University
Seoul, KoreaMar. 2019 - Feb. 2021

M.S. in Computer ScienceJongwoo Lim

Hanyang University
Seoul, KoreaMar. 2015 - Feb. 2019

B.S. in Computer Science

Yonsei University - Mirae Campus
Wonju, KoreaMar. 2014 - Aug. 2014

B.S. in Computer EngineeringWithdrawal

Awards

  • Google Machine Learning Challenge Korea 5th place2017
  • Top start-up certification SW Maestro 7th2017
  • ACM-ICPC Asia Daejeon Regional Contest 16th place2016

Publications

Object detection in fisheye lens environment

Jiun Bae, Jongwoo Lim

Image Processing and Image Understanding, 32nd Workshop, 2020

Object counting using object detection and re-identification in video sequence

Gitaek Kwon, Jiun Bae, Jongwoo Lim

Image Processing and Image Understanding, 32nd Workshop, 2020

Collaborative Training of Balanced Random Forests for Open Set Domain Adaptation

Jongbin Ryu, Jiun Bae, Jongwoo Lim

arXiv, abs/2002.03642, 2020

Skills

Languages

  • Python
  • C++
  • JavaScript
  • Kotlin

Frameworks

  • PyTorch
  • TensorFlow
  • OpenCV
  • NumPy
  • FastAPI
  • Spring
  • React

Tools

  • Git
  • Docker
  • Linux