Completed
SIFT: David Lowe, UBC
Class Central Classrooms beta
YouTube playlists curated by Class Central.
Classroom Contents
Scale-invariant Feature Transform (SIFT) - Lecture 5
Automatically move to the next video in the Classroom when playback concludes
- 1 SIFT: David Lowe, UBC
- 2 SIFT - Key Point Extraction
- 3 Advantages
- 4 Invariant Local Features
- 5 Steps for Extracting Key Points
- 6 Scale Space (Witkin, IJCAI 1983) • Apply whole spectrum of scales
- 7 Approximation of LOG by Difference of Gaussians
- 8 Building a Scale Space
- 9 How many scales per octave?
- 10 Initial value of sigma
- 11 Scale Space Peak Detection
- 12 Key Point Localization
- 13 Initial Outlier Rejection
- 14 Further Outlier Rejection
- 15 Orientation Assignment
- 16 Similarity to IT cortex
- 17 Extraction of Local Image Descriptors at Key Points
- 18 Descriptor Regions (n by n)
- 19 Key point matching
