Image classification using labelled and unlabelled data

In this paper we present a case study of co-training to image classification. We consider two scene classification tasks: indoors vs. outdoors and animals vs. sports. The results show that co-training with Naïve Bayes using 8-10 labelled examples obtained only 1.2-1.5% lower classification accuracy than Naïve Bayes trained on the full labelled version of the training set (138 examples in task 1 and 827 examples in task 2). Co-training was found to be sensitive to the choice of base classifier, with Naïve Bayes outperforming Random Forest. We also propose a simple co-training modification based on the different inductive basis of classification algorithms and show that it is a promising approach.