Material Classification in the Wild: Do Synthesized Training Data Generalise Better than Real-World Training Data?
Grigorios KalliatakisAnca SticlaruGeorge StamatiadisShoaib EhsanAles LeonardisJuergen GallKlaus D. McDonald-MaierPublished in: CoRR (2017)
Keyphrases
- training data
- real world
- data sets
- training set
- learning algorithm
- training process
- decision trees
- test set
- supervised learning
- training dataset
- test data
- training examples
- text categorization
- databases
- classification models
- data mining
- naive bayes
- unlabeled data
- prior knowledge
- support vector machine
- classification accuracy
- training samples
- probabilistic model
- labeled data
- input data
- case study
- generalization error
- neural network