SIFT Scale Space
The scale-invariant feature transform, developed by David Lowe and presented in 1999, finds distinctive keypoints that survive changes in scale and rotation. It underpinned panorama stitching and object recognition for years.
Spot it
- 01Circles scaled to features
- 02Orientation lines inside rings
- 03Stack of blurred octaves
Why it happens
Builds a stack of progressively blurred images, subtracts neighbors to form difference-of-Gaussian layers, and keeps points that are extrema across both space and scale.
Recreate it in Pixio
Image edit prompt“Visualize SIFT keypoints on this image: keep the photo at reduced brightness and overlay colored circles whose radius matches each feature's scale, with a line from the center showing dominant orientation. Small circles sit on fine detail, large circles on big structures. Use cyan and yellow. Keep the original subject and composition recognizable.”
Open PixioUpload your image and pick an image-edit model.