People Detection benchmark repository
Results
Author: Victor Fernandez-Carbajales .
Contact university/company: Universidad Autonoma de Madrid.
Contact email: victor.fernandez@uam.es
Method name: Fusion.
Description and pocessing time: The Fusion detector is a real time detection approach based on segmentation and a holistic person model. The initial objects candidates to be person are extracted using background subtraction and the holistic person model is the combination or fusion at decision level of three simple person models: ellipse fitting, ghost and aspect ratio. The Fusion approach is implemented in C++ (OpenCV) and the computational cost is around 0.02 seconds per frame with 352x288 images. The tests have been performed on a Pentium IV with a CPU frequency of 2.4 GHz and 3GB RAM.
Parameters: Author default parameters [1].
You can download the people detection results of this method by clicking here.
Results for each video sequence:
Video |
Fusion |
1 |
34.9 |
2 |
92.5 |
3 |
64.3 |
4 |
0.5 |
5 |
3.5 |
6 |
8.1 |
7 |
7.6 |
8 |
17.7 |
9 |
45.9 |
10 |
21.2 |
11 |
2.1 |
12 |
65.7 |
13 |
15.5 |
14 |
12.7 |
15 |
16.6 |
16 |
37.7 |
Average AUC |
27.9 |
Ranking |
5.69 |
Results for each background complexity:
Background complexity |
Fusion |
Baseline |
63.9 |
Dynamic Background |
2.0 |
Camera Jitter |
7.9 |
Intermittent Object Motion |
21.7 |
Shadow |
29.6 |
Average AUC |
25.0 |
Ranking |
6.80 |
Results for each classification complexity:
Classification complexity |
Fusion |
Low |
48.7 |
Medium |
23.1 |
High |
10.3 |
Average AUC |
27.4 |
Ranking |
6.33 |
Associated references:
[1] V. Fernandez-Carbajales, M. A. Garcia, and J. M. Martinez, “Robust people detection by fusion of evidence from multiple methods,” in Proc. of WIAMIS, 2008, pp. 55–58.
