Feature Detection

---

• Object Recognition from Local Scale-Invariant Features, David G. Lowe iccv99.pdf
• Distinctive image features from scale-invariant keypoints, David G. Lowe, International Journal of Computer Vision, 60, 2 (2004), pp. 91-110. ijcv04.pdf
• Wikipedia SIFT article, http://en.wikipedia.org/wiki/Scale-invariant_feature_transform
• http://homepages.inf.ed.ac.uk/rbf/CVonline/LOCAL_COPIES/AV0405/MURRAY/SIFT.html
• SIFT lecture http://www.cs.nccu.edu/~xhuang/cisg5740-f10/16-Sift.pdf Xinyu Huang and Sebastian Thrun's slides

---

• Feature definitions vary but essentially a salient part of the image or data
• Some also have a notion of a Feature descriptor or feature vector
• SIFT is a very popular example that uses many techniques

SIFT Feature Extraction Overview

1. Scale-space extrema detection: The first stage of computation searches over all scales and image locations. It is implemented efficiently by using a difference-of-Gaussian function to identify potential interest points that are invariant to scale and orientation.
2. Keypoint localization: At each candidate location, a detailed model is fit to determine location and scale. Keypoints are selected based on measures of their stability.
3. Orientation assignment: One or more orientations are assigned to each keypoint location based on local image gradient directions. All future operations are performed on image data that has been transformed relative to the assigned orientation, scale, and location for each feature, thereby providing invariance to these transformations.
4. Keypoint descriptor: The local image gradients are measured at the selected scale in the region around each keypoint. These are transformed in to a representation that allows for significant levels of local shape distortion and change in illumination