0

My understanding is that SIFT is made scale invariant by using different scales for the gaussian smoothing, why is an image pyramid also used?

Improve this question
4
  • what answers are you looking for that aren't addressed in the SIFT paper? what is your understanding of image pyramids? Commented Apr 22, 2022 at 12:21
  • My understanding is that the spread of the gaussian is used to produce scale invariance, I do not understand why both scale changes and an image pyramid is used(though some sources don't appear to say the image is down sampled). Commented Apr 22, 2022 at 21:41
  • because the pyramid reduces the amount of data to... sift through. sufficiently lowpassed data can be safely decimated with no loss in information (from the decimation). Commented Apr 23, 2022 at 11:26
  • So is the orientation and descriptor computed in the resampled image? Commented Apr 23, 2022 at 20:27
Related questions
6
How to use DoG Pyramid in SIFT
1
SIFT descriptors on subimages do not sum up to those on the whole image?
6
SIFT prediction
Load 7 more related questions Show fewer related questions

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.