Object detection using image reconstruction with PCA

1
Object detection using image reconstruction with
PCA

• Source Image and Vision Computing. March
• 2007.
• Authors Luis Malagon-Borja, Olac Fuentes.
• Reporter Yen-Chang, Chen.
• Date 2007/1/9.

2
Outline

• Introduction
• PCA classifier
• Adding a Support Vector Machine classifier
• Experimental results
• Conclusions

3
Introduction
The detection system
Output?0 gt Pedestrian Outputlt0 gt Non-pedestrian
Classifier based on Image Reconstruction with PCA
SVM classifier
Reduction of false detections by means of
heuristics
-Eliminating single detections -Eliminating
nearby detections
Input image
Output image
4
PCA classifier
The principal component.PC (The eigenvectors of C)
P The first k eigenvectors of PC.
Mean object of the set
Covariance matrix C
r The reconstructed image
p The projection of the sub-image u.
Sub-image u.
d Reconstruction error
5
PCA classifier
Ex
Ex
The sets
g e An image
6
Adding a Support Vector Machine classifier
Support vector
Support vector
7
Adding a Support Vector Machine classifier
Input image
SVM classifier
PCA classifier
(1)Group the detections
(2) Eliminating single detections
(1)Define a region
(2) Choice max Preference to keep
8
Experimental results
The capability of the system for detecting people
in still images with cluttered backgrounds.
9
Experimental results
ROC curves comparing the performance of out
classifiers versus the best reported in the
literature.
10
Conclusions

• Authors have presented an object detection system
  for static images.
• This system is able to detect frontal and rear
  views of pedestrians, and usually it can also
  detect side views of pedestrians.