Object Detection

Five questions. Choose one answer for each, then submit to see your score and explanations.

The practice target is 80%. You can retry as often as you like, and every lesson stays accessible.

You can practice as a guest. Sign in before starting an attempt to save your scores.

1. What does object detection predict?A. Object categories and their locationsB. Only a single category for the whole imageC. A precise class for every pixel by definitionD. Only camera settings

2. How is intersection over union defined?A. Overlap area divided by image widthB. The sum of box widthsC. The confidence score divided by the number of classesD. Overlap area divided by union area

3. What is non-maximum suppression mainly used for?A. Training a model without labelsB. Converting all detections into a segmentation maskC. Reducing duplicate overlapping detectionsD. Increasing the image resolution

4. Two boxes each have area 100, with overlap area 50. What is their IoU?A. OneB. One thirdC. One halfD. Two thirds

5. A detector returns 10 boxes: 8 true positives and 2 false positives. There are 16 ground-truth objects. Assuming correct one-to-one matching, what are precision and recall?A. 80% precision and 50% recallB. 50% precision and 80% recallC. 80% precision and 80% recallD. 20% precision and 50% recall

Review this module from its first lesson →

Object Detection

Your result