Data Annotation & LabelingBeginnerTier 1 · $25
Image Bounding Boxes
Draw tight, class-correct boxes for object detection datasets and defend them under QA review.
3 modules160 minutes of material10 exam questions · 80% to pass
Syllabus
01Box geometry and tightness
45 min- Place boxes that hug visible object extent
- Recognise over- and under-fitting
02Class rules, occlusion and truncation
60 min- Apply the label set when classes overlap
- Handle objects cut by the frame edge
03QA, reject reasons and appeals
55 min- Write a reject reason a reviewer can act on
- Appeal a QA decision with evidence
Assessed concepts
The exam tests these directly. Definitions are summarised here; the exact answer key stays on the server.
Tightness
Occlusion handling
Class boundary rule
Truncation
Reject policy