Skip to content
View ShvetsMaksym42's full-sized avatar

Block or report ShvetsMaksym42

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
ShvetsMaksym42/README.md

Hi there 👋

I'm Maksym, focused on applied Computer Vision and Machine Learning, based in Germany. I prioritize implementing core components to understand them deeply.

Tools & Tech

  • Language: Python
  • Deep Learning & CV: PyTorch, Albumentations, Ultralytics YOLOv8, Grad-CAM, TensorBoard
  • Core Libraries: OpenCV, NumPy, Pandas, Scikit-learn

Recent Projects:

  • Built an end-to-end steel defect detection pipeline using YOLOv8m — including RLE decoding, image tiling, bbox extraction, stratified splitting, model training, and evaluation Severstal Steel Defect Detection
  • Developed a lightweight custom CNN (~16x smaller than ResNet18) for industrial casting inspection, achieving comparable performance with significantly reduced model size, verified via Grad-CAM Industrial Casting Inspection.

Pinned Loading

  1. steel-surface-defect-detection steel-surface-defect-detection Public

    Steel surface defect detection using YOLOv8m on the Severstal dataset. Full pipeline: RLE-to-bbox conversion, image tiling, stratified splitting, custom balanced sampling, and Gradio deployment on …

    Jupyter Notebook 1

  2. industrial-casting-inspection industrial-casting-inspection Public

    Deep Learning system for industrial quality control: detecting casting defects in submersible pump impellers using a lightweight 4MB CNN.

    Python 2