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

TL;DR

This paper introduces Detransformation Autoencoder (DTAE), a self-supervised approach that enhances open set recognition by learning transformation-invariant representations, improving unknown detection and known class classification.

Contribution

The paper proposes a novel self-supervised autoencoder method, DTAE, that improves open set recognition by learning transformation-invariant features and outperforming existing methods.

Findings

Pre-training with DTAE improves OSR performance.

DTAE enhances unknown class detection accuracy.

Representations contain more class info and less transformation info.

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.