Skip to content
View weiykong's full-sized avatar

Block or report weiykong

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
weiykong/README.md

Hi, I'm Weiyuan Kong 👋

Computer Vision, Scientific Imaging & Applied AI Engineer

PhD in Biophysics · Image Analysis · Quantitative Microscopy · Scientific Software · AI-Assisted Workflows

LinkedIn Location


🔬 Overview

I build tools that turn complex scientific images into reliable, measurable data.

My background is in biophysics and quantitative imaging, where I worked on extracting mechanical and morphological information from microscopy experiments. Over time, I became increasingly interested in the engineering side of science: building robust pipelines, automating analysis workflows, validating results, and turning research methods into reusable software.

Today, I work mainly at the intersection of:

  • 🤖 Applied AI & automation
  • 🔬 Scientific imaging
  • 👁️ Computer vision
  • 📊 Quantitative analysis
  • 🛠 Reproducible software engineering

🛠 Tech Stack

Category Badges
Languages
Image Analysis & ML
Engineering & Tools

🚀 Featured Projects

Image dataset quality-control toolkit for computer vision workflows.
Detects duplicates, blur, corruption, exposure issues, and train/validation leakage. Exports HTML, JSON, and CSV reports.
Python OpenCV Data Quality Control Reports

Scientific imaging package for extracting mechanical information from microscopy images.
Includes topology extraction, Bayesian force inference, curvature analysis, and stress summaries.
Python Scientific-Computing Microscopy Bayesian

Android computational photography prototype combining depth estimation with bokeh rendering.
Combines monocular depth estimation with GPU-based bokeh rendering.
Android Kotlin GPU Shaders Computational Photography

Configurable PyTorch U-Net pipeline for biological image segmentation.
Streamlined training, inference, and validation pipelines for microscopy and biological image datasets.
PyTorch U-Net Image Segmentation Deep Learning


📈 GitHub Stats

Weiyuan's GitHub Stats

Stats


💡 Philosophy & Interests

"Good scientific software should not only work once. It should be understandable, testable, reusable, and useful to someone else."

  • Current Focus: Scientific imaging and quantitative microscopy, computer vision for biological/medical images, dataset quality control prior to ML training, inverse problems, and physics-informed image analysis.

Pinned Loading

  1. ForceInferencePy ForceInferencePy Public

    A robust Python package for force inference in 2D and 2.5D tissues.

    Python 2

  2. imgclean imgclean Public

    Audit and clean image datasets — find duplicates, blurry images, corrupted files, and train/val/test leakage before training or labeling.

    Python 2

  3. CineDepthPro CineDepthPro Public

    On-device AI portrait mode for Android. Monocular depth estimation + AGSL GPU bokeh engine with 7 lens effects.

    Kotlin 2

  4. universal-unet-segmentation universal-unet-segmentation Public

    A universal U-Net implementation designed for high-performance biological image segmentation. This project is optimized for TIFF stacks and supports acceleration on Apple Silicon (M1/M2/M3) as well…

    Python 1