Image Outlier Detection Without Training using RANSAC

TL;DR

This paper introduces RANSAC-NN, a training-free image outlier detection method that directly operates on datasets with outliers, improving robustness and simplifying the process compared to traditional trained models.

Contribution

The paper presents RANSAC-NN, a novel outlier detection algorithm that does not require training and can handle datasets with outliers directly, unlike existing methods.

Findings

RANSAC-NN performs favorably on benchmark datasets.

It enhances robustness when integrated with existing methods.

It eliminates the need for data examination and model training.

Abstract

Image outlier detection (OD) is an essential tool to ensure the quality of images used in computer vision tasks. Existing algorithms often involve training a model to represent the inlier distribution, and outliers are determined by some deviation measure. Although existing methods proved effective when trained on strictly inlier samples, their performance remains questionable when undesired outliers are included during training. As a result of this limitation, it is necessary to carefully examine the data when developing OD models for new domains. In this work, we present a novel image OD algorithm called RANSAC-NN that eliminates the need of data examination and model training altogether. Unlike existing approaches, RANSAC-NN can be directly applied on datasets containing outliers by sampling and comparing subsets of the data. Our algorithm maintains favorable performance compared to existing methods on a range of benchmarks. Furthermore, we show that RANSAC-NN can enhance the robustness of existing methods by incorporating our algorithm as part of the data preparation process.