Guide-linked quiz · Medium Level

Texture Versus Shape Bias in CNNs Quiz

Probe cue-conflict images, texture and shape decisions, stylized training, task-dependent cues, and why clean accuracy is insufficient.

Question 1 of 8

An image has a cat outline with a different animal’s texture. What can its classification reveal?

Preview every question8 questions

Answers and explanations appear as you take the quiz.

  1. An image has a cat outline with a different animal’s texture. What can its classification reveal?
    • Whether clean-set accuracy rose
    • Whether the camera has metric depth
    • Whether the model follows shape or texture when cues disagree
    • Whether the generated image has perfect color calibration
  2. Why does clean-image accuracy alone not reveal shape preference?
    • Shape and texture often agree on ordinary examples
    • Clean examples contain no silhouettes
    • Accuracy cannot be computed for CNNs
    • Texture and shape always conflict on clean data
  3. What did the Geirhos and colleagues study observe for the CNNs it evaluated?
    • Every CNN ignores global shape in all settings
    • People never rely on texture
    • Stylized training removes every failure mode
    • They often followed texture cues in the designed conflict tests
  4. When might texture use be the intended behavior rather than a flaw?
    • When identity should persist despite changed finish
    • When classifying materials or surface defects
    • When the target class is a geometric silhouette
    • When texture is deliberately unrelated to the label
  5. A model trained with stylized images shows more shape choices. What remains necessary?
    • It proves robustness to every natural shift
    • It makes clean-image accuracy unnecessary
    • Test on clean and target-domain photos for the actual job
    • It validates every material-recognition application
  6. Why should style-transfer artifacts be inspected in a cue-conflict set?
    • They may create a third shortcut unrelated to intended shape or texture
    • Artifacts can only improve human judgments
    • Artifacts are guaranteed independent of labels
    • The model cannot learn generated-image artifacts
  7. Which measure best describes a shape-bias score?
    • A direct measure of human scene understanding
    • The model’s overall clean accuracy
    • A guarantee against common corruptions
    • The rate of shape-consistent choices on a specified cue-conflict set
  8. Why might maximizing shape bias harm a fabric inspector?
    • Fabric defects contain no surface information
    • Surface texture may contain the defects the inspector must detect
    • Shape preference is always better for every task
    • Texture cues never survive camera changes

Keep testing yourself

More quizzes picked for your level.

View all quizzes