XINYU
🚧 This website is still under development

Hello, I am Xinyu.

I work on Computer Vision

About Me

CV Engineer by Technology. Storyteller by Passion.

Xinyu Tian portrait

CV Engineer · Content Creator · World Explorer

I'm a Physics MSc student at NUS, currently building computer vision systems for industrial inspection at an A*STAR lab. I specialize in segmentation, anomaly detection, and making AI actually work in production.

Beyond the code, I document my student life, travels, and thinking journey as a bilingual content creator.

🎓 NUS MSc Physics🔬 A*STAR Research💻 Computer Vision
🔬

Computer Vision

Segmentation, anomaly detection & industrial inspection — building real systems at an A*STAR IME lab.

🤖

AI Engineering

Turning research into production-ready, explainable AI solutions with a focus on reliability.

📝

Research & Thesis

NUS MSc of Physics for Technology, focused on applied deep learning for industrial imaging and computer vision.

News

  • 04/2026NEWCompleted Research Internship at A*STAR
  • 08/2025Joined A*STAR as Research Intern
  • 08/2025Started MSc Physics at NUS
  • 06/2025Graduated from SWJTU with BSc in Applied Physics
  • 09/2021Started Applied Physics BSc at SWJTU

Experience

Internships and Professional Experience

  • Built an end-to-end CNN-based defect detection pipeline for AOG identification on wafer SEM images.
  • Curated and pixel-annotated 288 high-resolution SEM images using CVAT to construct the training set.
  • Benchmarked classical intensity baseline, plain UNet, and ResNet34-UNet — improved IoU, Dice, and F1 under identical settings.
  • Deployed the model as an application prototype reporting defect count, location, area, and batch-level statistical charts.
Python
PyTorch
OpenCV
Docker
CVAT
ResNet
UNet

Skills

Technical Skills and Languages

Programming

Python
C
C++
MATLAB
JavaScript
TypeScript

Computer Vision & ML

PyTorch
TensorFlow
OpenCV
NumPy

Tools

Git
GitHub
Docker
Linux
Jupyter
Anaconda
LaTeX
CVAT

Languages

Mandarin · Native
English · Professional
Cantonese · Beginner
Japanese · Beginner

Projects

Selected Work and Research Highlights

SEM Image Defect Detection

Case Study

Semantic segmentation and defect detection for industrial SEM imagery, focusing on robustness and deployment-ready design.

Computer VisionPyTorchU-Net
View Details ->

RGBD Multimodal Object Recognition

Case Study

A data pipeline and dashboard for logistics tracking, KPI visualization, and operational insights.

Data AnalyticsPythonVisualization
View Details ->

NUS Life & Travel Vlogs

Case Study

Short-form video content documenting campus life, Singapore culture, and travel stories.

Content CreationVideoStorytelling
View Details ->

Education

Academic Background and Qualifications

  • Researching deep learning-based semantic segmentation and anomaly detection for industrial SEM imaging.
  • Developing end-to-end CV pipelines with PyTorch for deployment-ready prototypes.
  • Collaborating with research groups on AI-driven solutions for advanced manufacturing quality control.
Computer Vision
Deep Learning
Machine Learning
Python
PyTorch
OpenCV
National University of Singapore mobile gallery image 1
National University of Singapore mobile gallery image 2
National University of Singapore mobile gallery image 3
National University of Singapore mobile gallery image 4
National University of Singapore mobile gallery image 5
National University of Singapore mobile gallery image 6
National University of Singapore mobile gallery image 7
National University of Singapore mobile gallery image 8
National University of Singapore mobile gallery image 9
  • Completed coursework in optics, electromagnetism, quantum mechanics, and computational physics.
  • Conducted experimental research in laboratory settings, developing strong data analysis and instrumentation skills.
  • Graduated with honours, building a foundation for graduate-level research in applied sciences.
C
C++
Python
MATLAB
NumPy
Public Speaking
Southwest Jiaotong University mobile gallery image 1
Southwest Jiaotong University mobile gallery image 2
Southwest Jiaotong University mobile gallery image 3
Southwest Jiaotong University mobile gallery image 4
Southwest Jiaotong University mobile gallery image 5
Southwest Jiaotong University mobile gallery image 6
Southwest Jiaotong University mobile gallery image 7
Southwest Jiaotong University mobile gallery image 8
Southwest Jiaotong University mobile gallery image 9
Southwest Jiaotong University mobile gallery image 10
Southwest Jiaotong University mobile gallery image 11
Southwest Jiaotong University mobile gallery image 12
Southwest Jiaotong University mobile gallery image 13
Southwest Jiaotong University mobile gallery image 14
Southwest Jiaotong University mobile gallery image 15
Southwest Jiaotong University mobile gallery image 16
Southwest Jiaotong University mobile gallery image 17
Southwest Jiaotong University mobile gallery image 18

Contact

Let's Connect

I'm open to collaboration, internship opportunities, and research conversations in computer vision and AI. Reach out if you want to work together or discuss ideas.

Connect with me

LinkedIn
GitHub
YouTube
Instagram
X
Facebook
Threads
RedNote