Download presentation
Presentation is loading. Please wait.
Published byFlora Houston Modified over 9 years ago
1
Using Statistic-based Boosting Cascade Weilong Yang, Wei Song, Zhigang Qiao, Michael Fang 1
2
Haar-like feature generation ◦ Integral image AdaBoost (feature selection) ◦ Limitations Statistic-based boosting (StatBoost) Cascade structure Performance evaluation 2
3
The objective is to learn a sequence of best weak classifiers and the best combining weights Each weak classifier is constructed based on one feature (1D). 3
4
Weak classifier 3 Final classifier: linear combination of the weak classifiers Weights Increased Weak Classifier 2 Weak Classifier 1 4
5
5
6
It takes WEEKS to train! 6
7
Training of the weak classifier 7 Figure courtesy to Minh-Tri Phan, 2007
8
Train the weak classifiers using statistics Assume the feature values of each class are of normal distribution Non-car Car Optimal threshold Feature value 8 Figure courtesy to Minh-Tri Phan, 2007
9
Constant Time 9 Figure courtesy to Minh-Tri Phan, 2007
10
Motivation ◦ Increase detection performance ◦ Reduce computation time Key insight ◦ Use simpler classifiers to reject the majority of sub- windows ◦ Use more complex ones to achieve low false positive rates on the rest of them 10
11
Overall form: degenerate decision tree Reflects the fact: within any single image an overwhelming majority of sub-windows are negative [1] 11 All Sub-windows 123 Rejected Sub-windows Further Processing T TT F FF
12
False positive rate Detection rate 12
13
Three-layer cascade Each layer has 4, 11, and 73 weak classifiers, respectively 13
14
Number of car images: 500 14 (Courtesy to UIUC Image Database for Car Detection)
15
Number of non-car images: 500 15 (Courtesy to UIUC Image Database for Car Detection)
16
1 st Feature2 nd Feature3 rd Feature 16
17
17
18
18
19
19
20
20
21
Test results on 170 Images: 21
22
1. P. Viola and M. Jones, “Rapid Object Detection Using a Boosted Cascade of Simple Features,” In Proc. CVPR, 2001. 2. M. Pham, T. Cham, “Fast training and selection of Haar features using statistics in boosting-based face detection.” In Proc. ICCV 2007 22
Similar presentations
© 2024 SlidePlayer.com Inc.
All rights reserved.