Understanding Center Loss Based Network for Image Retrieval with Few Training Data

Pallabi Ghosh, Larry S. Davis; Proceedings of the European Conference on Computer Vision (ECCV) Workshops, 2018, pp. 0-0

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 crossentropy 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.