Tech World Software SchoolTWSS

from first concept to confident practice

CV

Help computers find patterns in pixels and video.

Start learning 16 lessons · first lesson about 15 min · free · no account needed
About this track

Explore how computers see - the ideas behind face unlock, photo search, medical imaging and self-driving cars.

You’ll start from pixels and color, transform and filter images, find edges and count objects, then learn how models classify, detect and segment - from nearest neighbors to convolutional networks and Vision Transformers - and how to evaluate them fairly. Along the way you’ll paint pixels, mix colors, play with real filter kernels, guess box overlaps and step through detection algorithms. Every lesson ends with Python that runs in your browser, and the capstone builds a working object counter.

Before you start
  • No prior computer vision experience needed.
  • Basic Python helps; examples introduce the code as you go.
  • A modern browser with JavaScript enabled.
Loading progress…
  1. Unit 1 · 0/3 lessons

    Image foundations

    Represent images as pixel grids, work with color, and transform images. Badge: Pixel reader

    1. 0x001Represent an image as dataSee that every image is a grid of numbers, and read its size and pixels. 15 min 0/2 exercises solved 15 min 0/2 exercises solved
    2. 0x102Work with color and image arraysMix colors from red, green and blue, convert them to gray, and mind the channel order. 18 min 0/2 exercises solved 18 min 0/2 exercises solved
    3. 0x203Transform images: flip, rotate, crop and resizeChange an image’s geometry and brightness by rearranging and adjusting its numbers. 20 min 0/2 exercises solved 20 min 0/2 exercises solved
  2. Unit 2 · 0/4 lessons

    Image operations

    Threshold, filter, find edges and count objects. Badge: Pattern spotter

    1. 0x304Separate foreground from backgroundTurn an image into a foreground mask with a threshold, and pick the threshold from a histogram. 20 min 0/2 exercises solved 20 min 0/2 exercises solved
    2. 0x405Detect patterns with image filtersSlide a small grid of weights over an image to blur, sharpen and find edges. 22 min 0/2 exercises solved 22 min 0/2 exercises solved
    3. 0x506Find edges with gradientsMeasure how fast brightness changes, combine directions, and outline objects. 20 min 0/1 exercises solved 20 min 0/1 exercises solved
    4. 0x607Count objects with connected componentsGroup touching foreground pixels into objects, then measure and box each one. 22 min 0/2 exercises solved 22 min 0/2 exercises solved
  3. Unit 3 · 0/3 lessons

    Learning to see

    Classify images from examples, build CNN intuition, and train with augmentation. Badge: Feature learner

    1. 0x708Classify images by learning from examplesSee why hand-written rules fail, then classify images by comparing them to labeled examples. 20 min 0/1 exercises solved 20 min 0/1 exercises solved
    2. 0x809Build intuition for convolutional networksStack learned filters and pooling to turn pixels into features, and work out each layer’s size. 22 min 0/2 exercises solved 22 min 0/2 exercises solved
    3. 0x9010Train vision models: data, augmentation and transferLearn how vision models are trained, why they overfit, and how augmentation and pretrained models help. 20 min 0/1 exercises solved 20 min 0/1 exercises solved
  4. Unit 4 · 0/3 lessons

    Find objects

    Choose the right task, measure box overlap, and remove duplicate detections. Badge: Vision builder

    1. 0xA011Know the vision tasks: classify, detect, segmentTell classification, detection and segmentation apart, and work with bounding-box formats. 18 min 0/1 exercises solved 18 min 0/1 exercises solved
    2. 0xB012Evaluate predictions and bounding boxesScore how well a predicted box matches the truth with IoU, and evaluate a detector. 22 min 0/2 exercises solved 22 min 0/2 exercises solved
    3. 0xC013Remove duplicate detections with non-maximum suppressionKeep one box per object by suppressing weaker boxes that overlap a stronger one. 20 min 0/1 exercises solved 20 min 0/1 exercises solved
  5. Unit 5 · 0/2 lessons

    Modern and responsible vision

    Meet Vision Transformers and image-text models, and evaluate systems fairly. Badge: Responsible builder

    1. 0xD014Meet modern vision: transformers, CLIP and generationSee how images become sequences of patches, and how vision now connects to language. 20 min 0/1 exercises solved 20 min 0/1 exercises solved
    2. 0xE015Build computer vision responsiblyMeasure performance across groups, and weigh privacy and misuse before deploying vision systems. 18 min 0/1 exercises solved 18 min 0/1 exercises solved
  6. Unit 6 · 0/1 lessons

    Capstone

    Build a working object counter from the pieces you learned. Badge: Vision engineer

    1. 0xF016Capstone: build an object counterCombine thresholding, connected components and filtering into a pipeline that counts and boxes objects. 35 min 0/1 exercises solved 35 min 0/1 exercises solved