Shih-Fang Chen

Peer Reviewer

Reviewed 40+ manuscripts, assessing originality, methodological soundness, and empirical validity.

• IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
• IEEE Transactions on Multimedia (TMM)
• IEEE Transactions on Circuits and Systems for Video Technology (TCSVT)
• International Conference on Learning Representations (ICLR)
• IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

Intelligent Logistics Data Collection System

Developed and deployed an industrial computer vision system using image restoration and semantic segmentation for stacked-substrate measurement.

Invention patent: TWI802843B (2023)
Utility model patent: TW M615811 (2021)

Computer Vision & Machine Learning Researcher

NYCU / Academia Sinica · Ph.D. research fellowship

• Led end-to-end research in adaptive video perception, from problem formulation and model design to implementation, training, evaluation, and open-source release.
• Developed methods for multimodal learning, self-supervised pretraining, online model editing, and automated visual prompting.
• Built scalable training pipelines using PEFT, distributed training, activation checkpointing, and mixed precision for 20M+ training samples under limited compute resources.
• Evaluated robustness and generalization on 1M+ test video samples across seven public benchmarks; analyzed training dynamics, inference behavior, and failure modes.
• Published three first-author papers at ICLR, IEEE TCSVT, and IEEE TMM as the sole student author, and discussed progress with researchers at NYCU, Academia Sinica, and Microsoft AI.

HPC GPU System Administrator

Vision and Learning Lab, NYCU

Administered 10 GPU servers supporting 30 users, overseeing deployment and operations while coordinating shared compute resources across concurrent research projects.

Research Assistant, Computer Vision

Department of Electrical Engineering, Yuan Ze University · Full-time

• Led development of real-time vision applications as the sole vision-system developer, including 3D interaction, pose training, pitching analysis, and banknote reconstruction.
• Built and deployed NVIDIA edge systems using depth estimation, pose estimation, motion analysis, and image reconstruction; delivered systems to industry partners and contributed to patented technologies and CVGIP research.

Graduate Student Researcher, Machine Vision

Department of Computer Science, Yuan Ze University · M.S. industry–academia research

• Owned end-to-end vision system development, translating industry requirements into research prototypes and production-ready systems.
• Designed and deployed a PCB stacked-substrate counting and measurement system, automating manual inspection, reducing annual operating costs, and contributing to patents.

Computer Vision Project Lead & Developer

Department of Computer Science, Chaoyang University of Technology

Led an undergraduate team to develop a vision-based autonomous drone for disaster monitoring, receiving a Bronze Award at the Ministry of Education Creative Software Applications Contest.

Team Lead & Programmer, Autonomous Robotics

National LoTung Industrial Vocational High School

Led a student team to develop an autonomous robot and its perception and decision logic, earning first place at the World Robot Olympiad Taipei regional competition.

Ph.D. Fellowships

Academia Sinica; Foxconn–NYCU.

Travel Grants

FAOS; National Science and Technology Council (NSTC); International Conference on Learning Representations (ICLR).

Phi Tau Phi Honor Society

Honorary member.

National Yang Ming Chiao Tung University

Ph.D. in Computer Science

Coursework GPA: 4.24 / 4.3 · Dissertation GPA: 4.3 / 4.3

• Published first-author papers at leading machine learning and computer vision venues.
• Fellowships: Academia Sinica, Foxconn–NYCU. Travel grants: FAOS, NSTC, ICLR.
• Dissertation: Rethinking Generic Object Tracking Toward Human-Level Perceptual Intelligence.
• Advisor: Yen-Yu Lin (NYCU). Co-advisors: Jun-Cheng Chen (Academia Sinica), I-Hong Jhuo (Microsoft AI).

Yuan Ze University

M.S. in Computer Science

GPA: 4.0 / 4.0

• Conducted industry-collaborative research as the sole software developer, delivering a patented computer vision system into production and generating substantial annual cost savings.
• Ranked first in the graduating class and inducted into the Phi Tau Phi Honor Society.
• Advisor: Ting-Ying Chien.

Chaoyang University of Technology

B.S. in Computer Science

First place in the Computer Science project competition

• Led an award-winning drone project, receiving a Bronze Award from Taiwan's Ministry of Education and presenting the work at the AIT 2017 conference.
• Received a full scholarship for an academic exchange in Australia and the school-wide City Government Honors Student Award.
• Later led a 17-member team across departments in a university-wide exhibition.

National LoTung Industrial Vocational High School

High School Diploma, Department of Computer Science

Led a team while developing autonomous robot software, winning first place at the World Robot Olympiad Taipei regional competition.