Understanding Center Loss Based Network For Image Retrieval With Few Training Data

Performance of convolutional neural network based image retrieval depends on the characteristics and statistics of the data being used for training. We show that for training datasets with a large number of classes but small number of images per class, the combination of cross-entropy loss and center loss works better than either of the losses alone. While cross-entropy loss tries to minimize misclassification of data, center loss minimizes the embedding space distance of each point in a class to its center, bringing together data-points belonging to the same class.