Bounding Box Anomaly Scoring for simple and efficient Out-of-Distribution detection

2026-03-23

AI Summary

This work addresses the unreliability of deep neural networks when encountering out-of-distribution (OOD) inputs by proposing a post-hoc OOD detection method based on bounding-box abstraction. The approach constructs compact, axis-aligned activation bounding boxes in feature space to effectively discriminate between in-distribution and out-of-distribution samples. It introduces a hierarchical anomaly scoring scheme, tailors monitoring variables for convolutional layers, and decouples the clustering and bounding-box construction mechanisms to balance expressive power with model simplicity. Experimental results demonstrate that the method achieves robust separation between in- and out-of-distribution data on standard image classification benchmarks, while maintaining a compact structure, computational efficiency, and support for online updates.

Technology Category

• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Computer Vision: Object Detection & Categorization
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text

Abstract

Out-of-distribution (OOD) detection aims to identify inputs that differ from the training distribution in order to reduce unreliable predictions by deep neural networks. Among post-hoc feature-space approaches, OOD detection is commonly performed by approximating the in-distribution support in the representation space of a pretrained network. Existing methods often reflect a trade-off between compact parametric models, such as Mahalanobis-based scores, and more flexible but reference-based methods, such as k-nearest neighbors. Bounding-box abstraction provides an attractive intermediate perspective by representing in-distribution support through compact axis-aligned summaries of hidden activations. In this paper, we introduce Bounding Box Anomaly Scoring (BBAS), a post-hoc OOD detection method that leverages bounding-box abstraction. BBAS combines graded anomaly scores based on interval exceedances, monitoring variables adapted to convolutional layers, and decoupled clustering and box construction for richer and multi-layer representations. Experiments on image-classification benchmarks show that BBAS provides robust separation between in-distribution and out-of-distribution samples while preserving the simplicity, compactness, and updateability of the bounding-box approach.

Problem

• Out-of-Distribution detection
• Bounding Box Anomaly Scoring
• feature-space approaches
• in-distribution support
• deep neural networks

Innovation

• Bounding Box Abstraction
• Out-of-Distribution Detection
• Post-hoc Anomaly Scoring
• Interval Exceedance
• Decoupled Clustering