Name
Abstract | ||
|---|---|---|
This paper proposes a method of feature co-occurrence representation based on boosting for object detection. A previously proposed method that combines multiple binary- classified codes by AdaBoost to represent the co-occurrence of features has been shown to be effective in face detec- tion. However, if an input feature is difficult to be assigned to a correct binary code due to occlusion or other factors, a problem arises here since the process of binary classifi- cation and co-occurrence representation may combine fea- tures that include an erroneous code. In response to this problem, this paper proposes a Co-occurrence Probabil- ity Feature (CPF) that combines multiple weak classifiers by addition and multiplication arithmetic operators using Real AdaBoost in which the outputs of weak classifiers are real values. Since CPF combines classifiers using two types of operators, diverse types of co-occurrence can be rep- resented and improved detection performance can be ex- pected. To represent even more diversified co-occurrence, this paper also proposes co-occurrence representation that applies a subtraction arithmetic operator. Although co- occurrence representation using addition and multiplica- tion operators can represent co-occurrence between fea- tures, use of the subtraction operator enables the represen- tation of co-occurrence between local features and features having other properties. This should have the effect of re- vising the probability of the detection-target class obtained from local features. Evaluation experiments have shown co- occurrence representation by the proposed methods to be effective. |
Year | DOI | Venue |
|---|---|---|
2010 | 10.1109/CVPRW.2010.5543173 | CVPR Workshops |
Field | DocType | Volume |
Computer vision,Object detection,AdaBoost,Binary classification,Pattern recognition,Computer science,Feature extraction,Multiplication,Artificial intelligence,Boosting (machine learning),Operator (computer programming),Face detection | Conference | 2010 |
Issue | ISSN | ISBN |
1 | 2160-7508 | 978-1-4244-7029-7 |
Citations | PageRank | References |
2 | 0.38 | 18 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Yuji Yamauchi | 1 | 43 | 10.45 |
| Masanari Takagi | 2 | 2 | 0.38 |
| Takayoshi Yamashita | 3 | 377 | 46.83 |
| fujiyoshi | 4 | 730 | 101.43 |