global-journal-of-computer-science-and-technology-d-neural-ai Global Journal of Computer Science and Technology - D: Neural & AI 0975-4350 0975-4172 Global Journals Publishing Group Incorporated 52846 Using Unlabeled Data for Increasing Low-Shot Classification Accuracy of Relevant and Open-Set Irrelevant Images Open-Set Low-Shot Classifier for Object Recognition KasapisSpiridon ZangGeng SmerekaJonathon M. VlahopoulosNickolas UNITED STATES, University of Michigan 26 05 2022 22 D2 11 24

In search, exploration, and reconnaissance tasks performed with autonomous ground vehicles, an image classification capability is needed for specifically identifying targeted objects (relevant classes) and at the same time recognize when a candidate image does not belong to anyone of the relevant classes (irrelevant images). In this paper, we present an open-set low-shot classifier that uses, during its training, a modest number (less than 40) of labeled images for each relevant class, and unlabeled irrelevant images that are randomly selected at each epoch of the training process. The new classifier is capable of identifying images from the relevant classes, determining when a candidate image is irrelevant, and it can further recognize categories of irrelevant images that were not included in the training (unseen). The proposed lowshot classifier can be attached as a top layer to any pre-trained feature extractor when constructing a Convolutional Neural Network.

 semi-supervised learning open-set classification neural networks receiver operating characteristic. Full Text

In search, exploration, and reconnaissance tasks performed with autonomous ground vehicles, an image classification capability is needed for specifically identifying targeted objects (relevant classes) and at the same time recognize when a candidate image does not belong to anyone of the relevant classes (irrelevant images). In this paper, we present an open-set low-shot classifier that uses, during its training, a modest number (less than 40) of labeled images for each relevant class, and unlabeled irrelevant images that are randomly selected at each epoch of the training process. The new classifier is capable of identifying images from the relevant classes, determining when a candidate image is irrelevant, and it can further recognize categories of irrelevant images that were not included in the training (unseen). The proposed low-shot classifier can be attached as a top layer to any pre-trained feature extractor when constructing a Convolutional Neural Network.