English

Self-supervised Detransformation Autoencoder for Representation Learning in Open Set Recognition

Machine Learning 2022-07-07 v2 Computer Vision and Pattern Recognition

Abstract

The objective of Open set recognition (OSR) is to learn a classifier that can reject the unknown samples while classifying the known classes accurately. In this paper, we propose a self-supervision method, Detransformation Autoencoder (DTAE), for the OSR problem. This proposed method engages in learning representations that are invariant to the transformations of the input data. Experiments on several standard image datasets indicate that the pre-training process significantly improves the model performance in the OSR tasks. Meanwhile, our proposed self-supervision method achieves significant gains in detecting the unknown class and classifying the known classes. Moreover, our analysis indicates that DTAE can yield representations that contain more target class information and less transformation information than RotNet.

Keywords

Cite

@article{arxiv.2105.13557,
  title  = {Self-supervised Detransformation Autoencoder for Representation Learning in Open Set Recognition},
  author = {Jingyun Jia and Philip K. Chan},
  journal= {arXiv preprint arXiv:2105.13557},
  year   = {2022}
}

Comments

arXiv admin note: text overlap with arXiv:2006.15117

R2 v1 2026-06-24T02:33:17.531Z