Image Segmentation

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. Which explanation matches semantic labels for every pixel?A. Instance segmentation separates object identities even if both share the same semantic class.B. Semantic segmentation assigns class labels per pixel without distinguishing individual objects of the same class.C. Intersection-over-union compares overlapping area, while boundary measures emphasize edge alignment.D. An annotation guide defines inclusion, occlusion, and boundary conventions.

2. Which explanation matches separate instances of the same class?A. Instance segmentation separates object identities even if both share the same semantic class.B. Semantic segmentation assigns class labels per pixel without distinguishing individual objects of the same class.C. Intersection-over-union compares overlapping area, while boundary measures emphasize edge alignment.D. An annotation guide defines inclusion, occlusion, and boundary conventions.

3. Which explanation matches measure mask overlap and boundary quality?A. Semantic segmentation assigns class labels per pixel without distinguishing individual objects of the same class.B. Instance segmentation separates object identities even if both share the same semantic class.C. An annotation guide defines inclusion, occlusion, and boundary conventions.D. Intersection-over-union compares overlapping area, while boundary measures emphasize edge alignment.

4. For separate instances of the same class, which result or action fits this practical case: Two tree canopies touch in the frame.A. Color all tree pixels with one tree class and compare with the reference mask.B. Report both IoU and branch-boundary errors on held-out masks.C. Give each tree its own mask before counting trees.D. Compare masks from two annotators, then resolve disagreements using the written rule.

5. For audit inconsistent mask annotations, which result or action fits this practical case: Two annotators trace the same tree canopy edge differently.A. Color all tree pixels with one tree class and compare with the reference mask.B. Compare masks from two annotators, then resolve disagreements using the written rule.C. Give each tree its own mask before counting trees.D. Report both IoU and branch-boundary errors on held-out masks.

Review this module from its first lesson →

Image Segmentation

Your result