← All workshopsRead Haar Cascades

Classical Machine Learning

Give a window two checks to pass

BOTIE

Total the bright and dark rectangles separately, subtract the dark total and require the contrast to exceed 2.
Ready to test
0 passed · 0 failed · 2 not run

An empty floor. Start with the information your machine needs.

Received result

Your answer will arrive here.

Examples, experiments and workspace controls
Live values

Connect a tool's inputs to see what it produces. You can try a branch before connecting it to Result.

Try your own inputs

Use a copy of the example to try your own data. Keep the problem's assumptions in mind, and predict the effect of a change before checking the output. You can also add another branch to compare two ways of doing the calculation.

Start at an output arrow, then click or drag to an input. Drag empty space to pan. Delete removes a selection. Ctrl/Cmd+Z undoes a change.
Context for this challenge

Total the bright and dark rectangles separately, subtract the dark total and require the contrast to exceed 2.

A face detector must examine many possible image windows. Simple checks can reject unlikely windows before more work is spent on them. Let's construct two checks from light-dark contrast and an edge response. Passing them makes a window a candidate; it does not establish that a face is present.

Inspect the component calculation
Strict threshold(
  Value: Subtract(
    Starting value: Add a list(
      Numbers: Bright rectangle pixels
    )
    Value to remove: Add a list(
      Numbers: Dark rectangle pixels
    )
  )
  Threshold: 2
)
Examples and results
Bright rectangle pixels
  1. 03
  2. 13
Dark rectangle pixels
  1. 01
  2. 11
Required result
1
Bright rectangle pixels
  1. 01
  2. 11
Dark rectangle pixels
  1. 01
  2. 11
Required result
0
Try a fresh construction

This clears this draft, including its undo history. Earlier successful attempts stay recorded. Build again without opening hints to record an independent attempt.

Stuck? Inspect a worked construction

The tools below form one possible solution. You can assemble it, follow its signals, and undo that change to return to your own construction.

Assumptions and test precision

This is a small construction problem. The lesson explains the assumptions and the parts of the full method that this example leaves out. Numerical tests allow a difference of 0.001. Lists and structures must also have the right shape.

Your last tested pieces can start the full construction. An existing draft there keeps its own pieces.

The second stage reaches its boundary

This check uses the pieces from “Require the window to survive both checks”.

BOTIE

The rectangle contrast passes its first-stage test. Predict the final cascade decision when the second-stage edge response is exactly one.
Bright rectangle pixels
  1. 02
  2. 12
  3. 22
Dark rectangle pixels
  1. 01
  2. 11
  3. 21
Second-stage edge response
1

Use a number, text, or an array matching the requested result. For a matrix, put each row in brackets. Saving your prediction runs your current construction on these inputs and checks the earlier examples too.

Build a construction in the workspace above first. This case will test those pieces.

Where could a deployed cascade stop work early, and why does passing these two hand-supplied tests still not prove a face is present?

Separate a cascade's rejection logic, its learned detector quality and its computational early-exit behaviour.

This is a small construction problem. The lesson explains the assumptions and the parts of the full method that this example leaves out.

Work through the Python challenges →
Continue the courseWhat Must a Model Improve On? →
← Image Alignment