Latent Log-Linear Models for Handwritten Digit Classification.
Thomas DeselaersTobias GassGeorg HeigoldHermann NeyPublished in: IEEE Trans. Pattern Anal. Mach. Intell. (2012)
Keyphrases
- handwritten digits
- log linear models
- log linear
- decision trees
- feature extraction
- training set
- pattern classification
- text classification
- handwritten chinese
- discriminative training
- character recognition
- mixture model
- image classification
- feature vectors
- object recognition
- support vector
- feature selection
- machine learning
- data sets
- supervised learning
- training samples
- class labels
- latent variables
- support vector machine