- Scale-space extrema detection: Search over multiple scales and image locations, the location and characteristic scales are given by DoG detector
- Keypoint localization: Select keypoints based on a measure of stability, keep only the strong interest points by eliminating the low-contrast and edge keypoints
- Orientation assignment: Compute the best orientation(s) for each keypoint region, which contributes to the stability of matching
- Keypoint descriptor computation: Use local image gradients at selected scale and rotation to describe each keypoint region
As discussed, SIFT is robust with regard to small variations in illumination (due to gradient and normalization), pose (small affine variation due to ...
