A method to make multiple hypotheses with high cumulative recognition rate using SVMs.
Kenichi MaruyamaMinoru MaruyamaHidetoshi MiyaoYasuaki NakanoPublished in: Pattern Recognit. (2004)
Keyphrases
- recognition rate
- support vector machine svm
- multiple hypotheses
- classification rate
- support vector machine
- multi class
- high recognition rate
- detection method
- rejection rate
- higher recognition rate
- segmentation method
- error rate
- training data
- level set
- classification accuracy
- spatial information
- dynamic programming
- face recognition
- image processing
- machine learning