Material Classification in theWild: Do Synthesized Training Data Generalise Better than Real-world Training Data?
Grigorios KalliatakisAnca SticlaruGeorge StamatiadisShoaib EhsanAles LeonardisJuergen GallKlaus D. McDonald-MaierPublished in: VISIGRAPP (4: VISAPP) (2018)
Keyphrases
- training data
- real world
- data sets
- decision trees
- training set
- test data
- training instances
- learning algorithm
- supervised learning
- training dataset
- prior knowledge
- training examples
- data mining
- wide range
- test set
- classification models
- class labels
- classification accuracy
- labeled data
- training samples
- neural network
- domain knowledge
- computer vision
- unlabeled data
- naive bayes
- input data
- synthetic data
- upper bound
- real life
- artificial neural networks
- training process
- case study
- sampled data
- unseen data
- learned from training data