update assets

pages/CSE559A/CSE559A_L17.md

@@ -166,10 +166,6 @@ Characteristic of good features:
 
 ##### Harris corner detector
 
-
-
-
-
 ### Applications of local features
 
 #### Image alignment

pages/CSE559A/CSE559A_L18.md

@@ -10,13 +10,13 @@ Goal: Descriptor distinctiveness
 Harris corner detector:
 
 > Other existing variants:
-> - Hessian & Harris: [Beaudet ‘78], [Harris ‘88]
+> - Hessian & Harris: [Beaudet '78], [Harris '88]
-> - Laplacian, DoG: [Lindeberg ‘98], [Lowe 1999]
+> - Laplacian, DoG: [Lindeberg '98], [Lowe 1999]
-> - Harris-/Hessian-Laplace: [Mikolajczyk & Schmid ‘01]
+> - Harris-/Hessian-Laplace: [Mikolajczyk & Schmid '01]
-> - Harris-/Hessian-Affine: [Mikolajczyk & Schmid ‘04]
+> - Harris-/Hessian-Affine: [Mikolajczyk & Schmid '04]
-> - EBR and IBR: [Tuytelaars & Van Gool ‘04]
+> - EBR and IBR: [Tuytelaars & Van Gool '04]
-> - MSER: [Matas ‘02]
+> - MSER: [Matas '02]
-> - Salient Regions: [Kadir & Brady ‘01]
+> - Salient Regions: [Kadir & Brady '01]
 > - Others…
 
 ### Deriving a corner detection criterion

pages/CSE559A/CSE559A_L19.md

@@ -43,15 +43,15 @@ $$
 \nabla^2_{norm}=\sigma^2\nabla^2\left(\frac{\partial^2}{\partial x^2}g+\frac{\partial^2}{\partial y^2}g\right)
 $$
 
-![Visualization of LoG](./images/Laplacian_of_Gaussian.png)
+![Visualization of LoG](https://notenextra.trance-0.com/CSE559A/Laplacian_of_Gaussian.png)
 
 #### Edge detection with LoG
 
-![Edge detection with LoG](./images/Edge_detection_with_LoG.png)
+![Edge detection with LoG](https://notenextra.trance-0.com/CSE559A/Edge_detection_with_LoG.png)
 
 #### Blob detection with LoG
 
-![Blob detection with LoG](./images/Blob_detection_with_LoG.png)
+![Blob detection with LoG](https://notenextra.trance-0.com/CSE559A/Blob_detection_with_LoG.png)
 
 ### Difference of Gaussians (DoG)
 

pages/CSE559A/CSE559A_L20.md

@@ -0,0 +1,64 @@
+# CSE559A Lecture 20
+
+## Local feature descriptors
+
+Detection: Identify the interest points
+
+Description: Extract vector feature descriptor surrounding each interest point.
+
+Matching: Determine correspondence between descriptors in two views
+
+### Image representation
+
+Histogram of oriented gradients (HOG)
+
+- Quantization
+  - Grids: fast but applicable only with few dimensions
+  - Clustering: slower but can quantize data in higher dimensions
+- Matching
+  - Histogram intersection or Euclidean may be faster
+  - Chi-squared often works better
+  - Earth mover’s distance is good for when nearby bins represent similar values
+
+#### SIFT vector formation
+
+Computed on rotated and scaled version of window according to computed orientation & scale
+
+- resample the window
+
+Based on gradients weighted by a Gaussian of variance half the window (for smooth falloff)
+
+4x4 array of gradient orientation histogram weighted by magnitude
+
+8 orientations x 4x4 array = 128 dimensions
+
+Motivation:  some sensitivity to spatial layout, but not too much.
+
+For matching:
+
+- Extraordinarily robust detection and description technique
+- Can handle changes in viewpoint
+  - Up to about 60 degree out-of-plane rotation
+- Can handle significant changes in illumination
+  - Sometimes even day vs. night
+- Fast and efficient—can run in real time
+- Lots of code available
+
+#### SURF
+
+- Fast approximation of SIFT idea
+- Efficient computation by 2D box filters & integral images
+  - 6 times faster than SIFT
+- Equivalent quality for object identification
+
+#### Shape context
+
+![Shape context descriptor](https://notenextra.trance-0.com/CSE559A/Shape_context_descriptor.png)
+
+#### Self-similarity Descriptor
+
+![Self-similarity descriptor](https://notenextra.trance-0.com/CSE559A/Self-similarity_descriptor.png)
+
+## Local feature matching
+
+### Matching