You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
These results are recorded from running a total of 35 experiments based on combinations of 7 detectors and 6 descriptors.
Performance Evaluation 1: Number of Keypoints
Count the number of keypoints on the preceding vehicle for all 10 images and take note of the distribution of their neighborhood size. Do this for all the detectors you have implemented.
Detector
Total keypoints from all 10 images
Total keypoints in scene (average per image)
Keypoints detected on preceding vehicle (average per image)
% keypoints removed from scene to focus on preceding vehicle
Shi-Tomasi
13423
1342
117
91.2166
HARRIS
1737
173
24
85.7225
FAST
17874
1787
149
91.6583
BRISK
27116
2711
276
89.8141
ORB
5000
500
116
76.78
AKAZE
13430
1343
167
87.5652
SIFT
13861
1386
138
90.0007
Performance Evaluation 2: Number of Matched Keypoints
Count the number of matched keypoints for all 10 images using all possible combinations of detectors and descriptors. In the matching step, the BF approach is used with the descriptor distance ratio set to 0.8.
Total number of matched keypoints over 10 images
Detector - Descriptor
BRISK
BRIEF
ORB
FREAK
AKAZE
SIFT
Shi-Tomasi
767
944
1814
766
0
928
HARRIS
142
173
320
146
0
163
FAST
899
1099
2162
881
0
1048
SIFT
300
339
0
270
0
800
AKAZE
1215
1266
2372
1188
2518
1274
ORB
751
545
1522
421
0
766
BRISK
1570
1704
3020
1526
0
1658
Average number of matched keypoints per image pair
Detector - Descriptor
BRISK
BRIEF
ORB
FREAK
AKAZE
SIFT
Shi-Tomasi
76
94
181
76
0
92
HARRIS
14
17
32
14
0
16
FAST
89
109
216
88
0
104
SIFT
30
33
0
27
0
80
AKAZE
121
126
237
118
251
127
ORB
75
54
152
42
0
76
BRISK
157
170
302
152
0
165
Performance Evaluation 3: Keypoint Detection and Descriptor Extraction
Log the time it takes for keypoint detection and descriptor extraction.