Skip to content
DeepForgeby svx
Beginnercomputer-visionconvolutionimage-processingfilters

Computer Vision Starter

28 problems·~11h·4 stages

Images as grids of numbers, then as features a model can use. You will implement grayscale conversion, padding, convolution, pooling, Gaussian blur, edge detection, and Otsu thresholding from scratch. A hands-on introduction for anyone curious how CNNs see.

Your progress

0/28 · 0%

Path readiness 0%Easy 0% · Medium 0% · Hard 0%

Continue path · RGB to Grayscale

What you will learn

  • Manipulate images as arrays: crop, pad, flip, threshold, and normalize
  • Build filters, blurring, histograms, and pooling from scratch
  • Detect edges and segment with morphology and Otsu's method
  • Connect classical vision to CNN and ViT building blocks

1Image Basics

0/8

Images are number grids. Convert color to grayscale, crop, pad, flip, invert, threshold, average-pool, and min-max normalize.

Stage checkpoint

0/2 to pass

A checkpoint is a short retrieval test drawn from this stage: solve 2 of 3 from memory to mark the stage complete. The stage stays open either way.

Tried this set and did not pass? Record it — the next set swaps in fresh problems from the same stage.
ProjectRecommended next stepBuild a CNN from ScratchHard

2Filtering and Blurring

0/8

Local operations: reflect padding, nearest-neighbour resize, cross-correlation, Gaussian kernels and blur, median filters, max pooling, and histogram binning.

Stage checkpoint

0/2 to pass

A checkpoint is a short retrieval test drawn from this stage: solve 2 of 3 from memory to mark the stage complete. The stage stays open either way.

Tried this set and did not pass? Record it — the next set swaps in fresh problems from the same stage.
ProjectRecommended next stepBuild a CNN from ScratchHard

3Morphology and Edges

0/6

Shape from pixels: dilation, erosion, bilinear interpolation, Otsu thresholding, full 2D convolution, and Sobel gradient magnitudes.

Stage checkpoint

0/2 to pass

A checkpoint is a short retrieval test drawn from this stage: solve 2 of 3 from memory to mark the stage complete. The stage stays open either way.

Tried this set and did not pass? Record it — the next set swaps in fresh problems from the same stage.
ProjectRecommended next stepBuild a CNN from ScratchHard

4Toward Recognition

0/6

From handcrafted features to learnt ones: histogram equalization, intersection over union, non-max suppression, and the convolution, pooling, and patch-embedding layers inside CNNs and ViTs.

Stage checkpoint

0/2 to pass

A checkpoint is a short retrieval test drawn from this stage: solve 2 of 3 from memory to mark the stage complete. The stage stays open either way.

Tried this set and did not pass? Record it — the next set swaps in fresh problems from the same stage.
LabRecommended next stepXOR BoundaryHard

Capstone

Collection

Close the path with one real artifact from the catalog.

Computer Vision Starter

Revisit the pixels, filtering, and edges stages as one image-processing warm-up.