Self-supervised Detransformation Autoencoder for Representation Learning in Open Set Recognition
It addresses the problem of open set recognition for image classification, offering an incremental improvement over existing methods like RotNet.
The paper tackles open set recognition by proposing a self-supervised Detransformation Autoencoder (DTAE) to learn invariant representations, resulting in significant performance gains in detecting unknown classes and classifying known ones on standard image datasets.
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.