A Unified Reasoning Framework for Holistic Zero-Shot Video Anomaly Analysis

TL;DR

This paper introduces a unified, zero-shot reasoning framework that integrates temporal detection, spatial localization, and semantic explanation for video anomaly analysis, enhancing interpretability and generalization without additional training.

Contribution

It presents a novel chained reasoning approach that connects multiple tasks in a zero-shot setting, improving holistic anomaly understanding in videos.

Findings

Achieves state-of-the-art zero-shot performance on multiple benchmarks.

Enhances interpretability through integrated spatial and semantic explanations.

Demonstrates effective reasoning power of foundation models in video analysis.

Abstract

Most video-anomaly research stops at frame-wise detection, offering little insight into why an event is abnormal, typically outputting only frame-wise anomaly scores without spatial or semantic context. Recent video anomaly localization and video anomaly understanding methods improve explainability but remain data-dependent and task-specific. We propose a unified reasoning framework that bridges the gap between temporal detection, spatial localization, and textual explanation. Our approach is built upon a chained test-time reasoning process that sequentially connects these tasks, enabling holistic zero-shot anomaly analysis without any additional training. Specifically, our approach leverages intra-task reasoning to refine temporal detections and inter-task chaining for spatial and semantic understanding, yielding improved interpretability and generalization in a fully zero-shot manner. Without any additional data or gradients, our method achieves state-of-the-art zero-shot performance across multiple video anomaly detection, localization, and explanation benchmarks. The results demonstrate that careful prompt design with task-wise chaining can unlock the reasoning power of foundation models, enabling practical, interpretable video anomaly analysis in a fully zero-shot manner. Project Page: https://rathgrith.github.io/Unified_Frame_VAA/.