en.wikipedia.org/w/index.php?title=Bitter_lesson&oldid=1310091576
1 correction found
detecting high-level features with SIFT
SIFT does not detect high-level features. It is a method for detecting and describing local keypoints/local features in images.
Full reasoning
This phrase misdescribes what SIFT does.
David Lowe’s own SIFT materials describe it as a "keypoint detector" and point readers to the paper "Distinctive image features from scale-invariant keypoints." The standard description of SIFT is that it extracts local scale-invariant features/keypoints and computes descriptors around them.
Likewise, Scholarpedia’s overview of SIFT describes it as an image descriptor for matching and recognition based on collections of local image features. That contradicts the article’s wording: SIFT is not a method for detecting high-level semantic features.
So the technically correct description is that SIFT detects/describes local keypoints or local invariant features, not high-level features.
2 sources
- Keypoint detector
The SIFT Keypoint Detector ... The most complete and up-to-date reference for the SIFT feature detector is given in the following journal paper: David G. Lowe, "Distinctive image features from scale-invariant keypoints".
- Scale Invariant Feature Transform - Scholarpedia
Scale Invariant Feature Transform (SIFT) is an image descriptor for image-based matching and recognition... Lowe demonstrated that robust and efficient recognition of objects in natural scenes can be performed based on collections of local image features.