A Survey on Open Set Recognition
Authors:
- 1 December 2021
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 37-44
Abstract
Open Set Recognition (OSR) is about dealing with unknown situations that were not learned by the models during training. In this paper, we provide a survey of existing works about OSR and distinguish their respective advantages and disadvantages to help out new researchers interested in the subject. The categorization of OSR models is provided along with an extensive summary of recent progress. Additionally, the relationships between OSR and its related tasks including multi-class classification and novelty detection are analyzed. It is concluded that OSR can appropriately deal with unknown instances in the real-world where capturing all possible classes in the training data is not practical. Lastly, some new directions for future research topics are suggested.Keywords
All Related Versions
This publication has 26 references indexed in Scilit:
- Nearest neighbors distance ratio open-set classifierMachine Learning, 2016
- Sparse Representation-Based Open Set RecognitionIEEE Transactions on Pattern Analysis and Machine Intelligence, 2016
- SSD: Single Shot MultiBox Detector2016
- Collaborative Filtering Bandits2016
- Towards Open Set Deep Networks2016
- Open set recognition for automatic target classification with rejectionIEEE Transactions on Aerospace and Electronic Systems, 2016
- A bounded neural network for open set recognition2015
- Towards Open World Recognition2015
- Probability Models for Open Set RecognitionIEEE Transactions on Pattern Analysis and Machine Intelligence, 2014
- Toward Open Set RecognitionIEEE Transactions on Pattern Analysis and Machine Intelligence, 2012