← Back to curriculum

Module 5 — Segmentation & instance masks

Module 5 quiz & review

20 interactive multiple-choice questions on U-Net, Mask R-CNN, losses, and mIoU with lesson review links.

~45 min read + exercises

Before we begin

Aim for at least 16 out of 20 before the segmentation project.


Multiple choice quiz

Interactive quiz

Pick one answer per question. Feedback appears immediately — take your time before clicking.

0 / 20 correct·0 answered
  1. Question 1 of 20

    Semantic segmentation assigns:

    Answer options for question 1
  2. Question 2 of 20

    U-Net skip connections copy encoder features to the decoder to:

    Answer options for question 2
  3. Question 3 of 20

    Instance segmentation differs from semantic segmentation by:

    Answer options for question 3
  4. Question 4 of 20

    Dice loss is popular for segmentation when:

    Answer options for question 4
  5. Question 5 of 20

    Mean IoU (mIoU) for segmentation averages:

    Answer options for question 5
  6. Question 6 of 20

    Encoder–decoder architecture in segmentation:

    Answer options for question 6
  7. Question 7 of 20

    ROI Align in Mask R-CNN improves over ROI Pool by:

    Answer options for question 7
  8. Question 8 of 20

    Focal loss modifies cross-entropy to:

    Answer options for question 8
  9. Question 9 of 20

    Panoptic segmentation combines:

    Answer options for question 9
  10. Question 10 of 20

    Boundary F-score in segmentation evaluation:

    Answer options for question 10
  11. Question 11 of 20

    Transposed convolution (deconv) in decoders:

    Answer options for question 11
  12. Question 12 of 20

    Mask R-CNN adds to Faster R-CNN:

    Answer options for question 12
  13. Question 13 of 20

    Per-pixel cross-entropy segmentation loss with class weights:

    Answer options for question 13
  14. Question 14 of 20

    DeepLab atrous (dilated) convolutions enlarge receptive field by:

    Answer options for question 14
  15. Question 15 of 20

    Ignoring void/unlabeled pixels in mIoU means:

    Answer options for question 15
  16. Question 16 of 20

    Compared to classification, segmentation training labels are:

    Answer options for question 16
  17. Question 17 of 20

    Mask IoU matching in instance evaluation requires:

    Answer options for question 17
  18. Question 18 of 20

    Combining CE and Dice loss:

    Answer options for question 18
  19. Question 19 of 20

    SegFormer and transformer segmentation encoders use:

    Answer options for question 19
  20. Question 20 of 20

    Qualitative segmentation review should include:

    Answer options for question 20

What's next

Project: U-Net pet segmentation