FineVAU: A Novel Human-Aligned Benchmark for Fine-Grained Video Anomaly Understanding

TL;DR

This paper introduces FineVAU, a new benchmark for detailed video anomaly understanding, featuring a novel evaluation metric and dataset to better assess human-aligned, fine-grained analysis of unusual video events.

Contribution

The paper proposes a comprehensive benchmark with a new human-aligned evaluation metric and a high-quality dataset to improve fine-grained video anomaly understanding.

Findings

FVScore aligns better with human perception than existing metrics.

LVLMs struggle with spatial and temporal details in anomalies.

FineVAU reveals limitations of current models in detailed anomaly comprehension.

Abstract

Video Anomaly Understanding (VAU) is a novel task focused on describing unusual occurrences in videos. Despite growing interest, the evaluation of VAU remains an open challenge. Existing benchmarks rely on n-gram-based metrics (e.g., BLEU, ROUGE-L) or LLM-based evaluation. The first fails to capture the rich, free-form, and visually grounded nature of LVLM responses, while the latter focuses on assessing language quality over factual relevance, often resulting in subjective judgments that are misaligned with human perception. In this work, we address this issue by proposing FineVAU, a new benchmark for VAU that shifts the focus towards rich, fine-grained and domain-specific understanding of anomalous videos. We formulate VAU as a three-fold problem, with the goal of comprehensively understanding key descriptive elements of anomalies in video: events (What), participating entities (Who) and location (Where). Our benchmark introduces a) FVScore, a novel, human-aligned evaluation metric that assesses the presence of critical visual elements in LVLM answers, providing interpretable, fine-grained feedback; and b) FineW3, a novel, comprehensive dataset curated through a structured and fully automatic procedure that augments existing human annotations with high quality, fine-grained visual information. Human evaluation reveals that our proposed metric has a superior alignment with human perception of anomalies in comparison to current approaches. Detailed experiments on FineVAU unveil critical limitations in LVLM's ability to perceive anomalous events that require spatial and fine-grained temporal understanding, despite strong performance on coarse grain, static information, and events with strong visual cues.