-
Notifications
You must be signed in to change notification settings - Fork 0
/
index.html
44 lines (40 loc) · 2.42 KB
/
index.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
<!DOCTYPE html>
<html lang="en-us">
<head>
<meta charset="UTF-8">
<title>Cs482 hw3 by konst4</title>
<meta name="viewport" content="width=device-width, initial-scale=1">
<link rel="stylesheet" type="text/css" href="stylesheets/normalize.css" media="screen">
<link href='https://fonts.googleapis.com/css?family=Open+Sans:400,700' rel='stylesheet' type='text/css'>
<link rel="stylesheet" type="text/css" href="stylesheets/stylesheet.css" media="screen">
<link rel="stylesheet" type="text/css" href="stylesheets/github-light.css" media="screen">
</head>
<body>
<section class="page-header">
<h1 class="project-name">Cs482 hw3</h1>
<h2 class="project-tagline"></h2>
</section>
<section class="main-content">
<h3>
<a id="welcome-to-github-pages" class="anchor" href="#welcome-to-github-pages" aria-hidden="true"><span aria-hidden="true" class="octicon octicon-link"></span></a>CS482 _HW3 Konstantin Oleinikov</h3>
<p>Keypoints:</br>
I used Sift to detect my keypoints. I used Sift since I think it was more simple to understand. Basically in order to detect features the different algorithms look for corners in a image. I choosed for the image to be gray since it seems to be generally much simpler to work with. In order to get the orientation of the Different keypoints I had to set a flag for the draw keypoints function.
</br>
Matching:</br>
For matching I used Sift to detect the keypoints. Then I used the flannBasedMatcher instead of the provided bruteforce matcher. This matcher operates faster since it does not generally have to go through all the points one by one. Instead I believe it goes through nLog(n) points.
I choose to keep the threshold as is.If the threshold was increased more keypoints would be found. I choose to keep it at a minimum.
</br>
All Matching:</br>
My program goes through each permutation of the different images and calculates the # selected matches based on the threshold.It puts the result into a matrix and then saves it to a txt file using np.
</p>
<h3>
<a id="designer-templates" class="anchor" href="#designer-templates" aria-hidden="true"><span aria-hidden="true" class="octicon octicon-link"></span></a>Images</h3>
<img src="matching 4001_4053.PNG">
<img src="matching4001_4013.PNG">
<img src="matching4001_4023.PNG">
<img src="matching4001_4056.PNG">
<img src="matching4001_40005.PNG">
<h3>Keypoints</h3>
<img src="sift_keypoints.jpg">
</body>
</html>