Knowledge-Guided Textual Reasoning for Explainable Video Anomaly Detection via LLMs

TL;DR

This paper proposes a novel language-based framework for weakly supervised video anomaly detection that uses textual representations and knowledge reasoning to improve interpretability and reliability in surveillance applications.

Contribution

It introduces a new text-based approach that transforms videos into language, structures semantic knowledge, and generates explanations, enhancing interpretability over traditional visual feature methods.

Findings

Achieves competitive anomaly detection performance on UCF-Crime and XD-Violence datasets.

Provides interpretable, knowledge-grounded explanations for detected anomalies.

Demonstrates the effectiveness of textual reasoning in real-world surveillance scenarios.

Abstract

We introduce Text-based Explainable Video Anomaly Detection (TbVAD), a language-driven framework for weakly supervised video anomaly detection that performs anomaly detection and explanation entirely within the textual domain. Unlike conventional WSVAD models that rely on explicit visual features, TbVAD represents video semantics through language, enabling interpretable and knowledge-grounded reasoning. The framework operates in three stages: (1) transforming video content into fine-grained captions using a vision-language model, (2) constructing structured knowledge by organizing the captions into four semantic slots (action, object, context, environment), and (3) generating slot-wise explanations that reveal which semantic factors contribute most to the anomaly decision. We evaluate TbVAD on two public benchmarks, UCF-Crime and XD-Violence, demonstrating that textual knowledge reasoning provides interpretable and reliable anomaly detection for real-world surveillance scenarios.