A short quiz on vision system evaluation

Adrian F. Clark

  1. Which corner of a ROC curve indicates the best performance?

  2. What are the axes of a ROC curve?

  3. You are developing a automatic passport system for use by immigration, where pictures of people are compared to those in their passports. Which of the following is the best approach to take?

  4. What is `ground truth'?

  5. When evaluating vision systems, it is normal to:

  6. Which test is most appropriate for comparing algorithms' performances?

  7. What assumption underlies a null hypothesis test?

  8. What do we do if we want to see if algorithms' performances differ?

  9. What do we do if we want to see if one algorithm's performance is better than another's ?

  10. What is a false negative?