Few-shot Open-set Recognition by Transformation Consistency

TL;DR

This paper introduces SnaTCHer, a novel method for few-shot open-set recognition that detects unseen classes without relying on pseudo-unseen samples, by measuring transformation consistency between prototypes.

Contribution

The paper proposes a transformation consistency-based detector, SnaTCHer, that improves open-set detection in few-shot learning without pseudo-unseen data dependence.

Findings

SnaTCHer outperforms existing methods in unseen class detection.

The method is robust across various prototype transformation techniques.

It enhances open-set recognition without sacrificing closed-set classification accuracy.

Abstract

In this paper, we attack a few-shot open-set recognition (FSOSR) problem, which is a combination of few-shot learning (FSL) and open-set recognition (OSR). It aims to quickly adapt a model to a given small set of labeled samples while rejecting unseen class samples. Since OSR requires rich data and FSL considers closed-set classification, existing OSR and FSL methods show poor performances in solving FSOSR problems. The previous FSOSR method follows the pseudo-unseen class sample-based methods, which collect pseudo-unseen samples from the other dataset or synthesize samples to model unseen class representations. However, this approach is heavily dependent on the composition of the pseudo samples. In this paper, we propose a novel unknown class sample detector, named SnaTCHer, that does not require pseudo-unseen samples. Based on the transformation consistency, our method measures the difference between the transformed prototypes and a modified prototype set. The modified set is composed by replacing a query feature and its predicted class prototype. SnaTCHer rejects samples with large differences to the transformed prototypes. Our method alters the unseen class distribution estimation problem to a relative feature transformation problem, independent of pseudo-unseen class samples. We investigate our SnaTCHer with various prototype transformation methods and observe that our method consistently improves unseen class sample detection performance without closed-set classification reduction.