Global k-NN Classifier for Small Sample Size Problems.

      Аннотация: What is Small Sample Size Problem? The methods need training samples to learn the information of the data. Some kind of data is high-dimensional, but no enough samples can be collected. When the number of training samples is much smaller than the dimension of data, we called this Small Sample Size Problem. In the face recognition field, small sample size problems generally exist.Let's talk about how to solve this problem better.

Ключевые слова: Computer Vision, Machine Learning, pattern classification