Week 06 · overview
Week 6: Detection, Segmentation, and Localization Evidence
Classification names an image-level category. Detection adds object location with boxes. Segmentation assigns classes at pixel level. More detailed output does not automatically mean more accurate or safer output.
| Day | Role | Evidence |
|---|---|---|
| 1 | Lesson 1: Object detection | Trace class, confidence, and box coordinates |
| 2 | Lesson 2: Segmentation | Distinguish semantic and instance masks |
| 3 | Investigation | Calculate overlap and expose localization errors |
| 4 | Lab | Compare box and mask intersection-over-union |
| 5 | Assessment + Reflection | Repair a “localized correctly” claim |
Your Localization Evidence Record preserves output type, target and prediction areas, intersection, union, IoU, claim repair, and limitation. The Lab uses supplied fictional annotations, activates no camera, and contacts no service.
The complete Microsoft repository and credits remain preserved in upstream/.
Review the source record.
Chapter question
How does Detection, Segmentation, and Localization Evidence change what evidence a student should trust?
What you will understand
You will connect the named mechanism to the evidence it can and cannot support.
What you will do
You will inspect supplied cases, trace the important calculation, and test one bounded claim.
What you will produce
You will produce the week evidence record named in the Lab or reflection.
How the week connects
This week carries the previous AI mechanism forward into a stricter evidence check.
Key vocabulary
mechanism; evidence; boundary; test; claim
Approximate time
Five class meetings at about 210 instructional minutes total.
Final evidence required
Submit the completed week record with at least two exact labels from the artifact.