Open-set Recognition via Augmentation-based Similarity Learning

TL;DR

This paper introduces OPG, a novel open set recognition method that uses augmentation-based pseudo-unseen data to learn similarity measures, effectively distinguishing known from unknown classes in open set scenarios.

Contribution

The paper proposes a new approach that generates pseudo-unseen data through distribution shifting augmentations to improve open set recognition performance.

Findings

Successfully distinguishes seen from unseen classes in benchmark datasets.

Outperforms existing methods in open set recognition tasks.

Demonstrates the effectiveness of similarity learning with pseudo-unseen data.

Abstract

The primary assumption of conventional supervised learning or classification is that the test samples are drawn from the same distribution as the training samples, which is called closed set learning or classification. In many practical scenarios, this is not the case because there are unknowns or unseen class samples in the test data, which is called the open set scenario, and the unknowns need to be detected. This problem is referred to as the open set recognition problem and is important in safety-critical applications. We propose to detect unknowns (or unseen class samples) through learning pairwise similarities. The proposed method works in two steps. It first learns a closed set classifier using the seen classes that have appeared in training and then learns how to compare seen classes with pseudo-unseen (automatically generated unseen class samples). The pseudo-unseen generation is carried out by performing distribution shifting augmentations on the seen or training samples. We call our method OPG (Open set recognition based on Pseudo unseen data Generation). The experimental evaluation shows that the learned similarity-based features can successfully distinguish seen from unseen in benchmark datasets for open set recognition.