Skip to content
#

convolution-from-scratch

Star

Here is 1 public repository matching this topic...

HAYDARKILIC / computer_vision

Star 0

A research-grade masterclass rebuilding the entire Computer Vision stack from first principles. No OpenCV, no PyTorch—just pure math and NumPy. Implements custom convolutions, Canny edge detection, eigenvalue-based Harris corners, Lucas–Kanade optical flow, and a CNN layer with full backward passes from scratch.

computer-vision image-processing cnn neural-networks optical-flow mathematical feature-matching cnn-from-scratch canny-edge-detector classical-cv convolution-from-scratch
  • Updated May 27, 2026
  • Jupyter Notebook

Improve this page

Add a description, image, and links to the convolution-from-scratch topic page so that developers can more easily learn about it.

Curate this topic

Add this topic to your repo

To associate your repository with the convolution-from-scratch topic, visit your repo's landing page and select "manage topics."

Learn more