Uncertainty-Weighted Image-Event Multimodal Fusion for Video Anomaly Detection

TL;DR

This paper introduces IEF-VAD, a novel framework that synthesizes event representations from RGB videos and fuses them with image features using an uncertainty-aware process, significantly improving video anomaly detection without dedicated event sensors.

Contribution

The paper presents a new multimodal fusion method that synthesizes event data from RGB videos and employs an uncertainty-aware approach for robust anomaly detection.

Findings

Sets new state-of-the-art on multiple benchmarks

Effectively emphasizes motion cues in RGB videos

Operates without dedicated event sensors

Abstract

Most existing video anomaly detectors rely solely on RGB frames, which lack the temporal resolution needed to capture abrupt or transient motion cues, key indicators of anomalous events. To address this limitation, we propose Image-Event Fusion for Video Anomaly Detection (IEF-VAD), a framework that synthesizes event representations directly from RGB videos and fuses them with image features through a principled, uncertainty-aware process. The system (i) models heavy-tailed sensor noise with a Student`s-t likelihood, deriving value-level inverse-variance weights via a Laplace approximation; (ii) applies Kalman-style frame-wise updates to balance modalities over time; and (iii) iteratively refines the fused latent state to erase residual cross-modal noise. Without any dedicated event sensor or frame-level labels, IEF-VAD sets a new state of the art across multiple real-world anomaly detection benchmarks. These findings highlight the utility of synthetic event representations in emphasizing motion cues that are often underrepresented in RGB frames, enabling accurate and robust video understanding across diverse applications without requiring dedicated event sensors. Code and models are available at https://github.com/EavnJeong/IEF-VAD.