k-NNN: Nearest Neighbors of Neighbors for Anomaly Detection

TL;DR

This paper introduces k-NNN, a novel operator that enhances anomaly detection algorithms by considering neighbors of neighbors in feature space, leading to improved detection across various datasets.

Contribution

The paper proposes the k-NNN operator that improves existing anomaly detection methods by incorporating second-order neighbor information in the embedding space.

Findings

Improved anomaly detection accuracy on diverse datasets

Enhancement achieved by replacing nearest neighbor with k-NNN in existing algorithms

Applicable to both homogeneous and diverse datasets

Abstract

Anomaly detection aims at identifying images that deviate significantly from the norm. We focus on algorithms that embed the normal training examples in space and when given a test image, detect anomalies based on the features distance to the k-nearest training neighbors. We propose a new operator that takes into account the varying structure & importance of the features in the embedding space. Interestingly, this is done by taking into account not only the nearest neighbors, but also the neighbors of these neighbors (k-NNN). We show that by simply replacing the nearest neighbor component in existing algorithms by our k-NNN operator, while leaving the rest of the algorithms untouched, each algorithms own results are improved. This is the case both for common homogeneous datasets, such as flowers or nuts of a specific type, as well as for more diverse datasets