Just Dance with $π$! A Poly-modal Inductor for Weakly-supervised Video Anomaly Detection

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Majhi, Snehashis, D'Amicantonio, Giacomo, Dantcheva, Antitza, Kong, Quan, Garattoni, Lorenzo, Francesca, Gianpiero, Bondarev, Egor, Bremond, Francois
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909615833219072
author Majhi, Snehashis
D'Amicantonio, Giacomo
Dantcheva, Antitza
Kong, Quan
Garattoni, Lorenzo
Francesca, Gianpiero
Bondarev, Egor
Bremond, Francois
author_facet Majhi, Snehashis
D'Amicantonio, Giacomo
Dantcheva, Antitza
Kong, Quan
Garattoni, Lorenzo
Francesca, Gianpiero
Bondarev, Egor
Bremond, Francois
contents Weakly-supervised methods for video anomaly detection (VAD) are conventionally based merely on RGB spatio-temporal features, which continues to limit their reliability in real-world scenarios. This is due to the fact that RGB-features are not sufficiently distinctive in setting apart categories such as shoplifting from visually similar events. Therefore, towards robust complex real-world VAD, it is essential to augment RGB spatio-temporal features by additional modalities. Motivated by this, we introduce the Poly-modal Induced framework for VAD: "PI-VAD", a novel approach that augments RGB representations by five additional modalities. Specifically, the modalities include sensitivity to fine-grained motion (Pose), three dimensional scene and entity representation (Depth), surrounding objects (Panoptic masks), global motion (optical flow), as well as language cues (VLM). Each modality represents an axis of a polygon, streamlined to add salient cues to RGB. PI-VAD includes two plug-in modules, namely Pseudo-modality Generation module and Cross Modal Induction module, which generate modality-specific prototypical representation and, thereby, induce multi-modal information into RGB cues. These modules operate by performing anomaly-aware auxiliary tasks and necessitate five modality backbones -- only during training. Notably, PI-VAD achieves state-of-the-art accuracy on three prominent VAD datasets encompassing real-world scenarios, without requiring the computational overhead of five modality backbones at inference.
format Preprint
id arxiv_https___arxiv_org_abs_2505_13123
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Just Dance with $π$! A Poly-modal Inductor for Weakly-supervised Video Anomaly Detection
Majhi, Snehashis
D'Amicantonio, Giacomo
Dantcheva, Antitza
Kong, Quan
Garattoni, Lorenzo
Francesca, Gianpiero
Bondarev, Egor
Bremond, Francois
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Weakly-supervised methods for video anomaly detection (VAD) are conventionally based merely on RGB spatio-temporal features, which continues to limit their reliability in real-world scenarios. This is due to the fact that RGB-features are not sufficiently distinctive in setting apart categories such as shoplifting from visually similar events. Therefore, towards robust complex real-world VAD, it is essential to augment RGB spatio-temporal features by additional modalities. Motivated by this, we introduce the Poly-modal Induced framework for VAD: "PI-VAD", a novel approach that augments RGB representations by five additional modalities. Specifically, the modalities include sensitivity to fine-grained motion (Pose), three dimensional scene and entity representation (Depth), surrounding objects (Panoptic masks), global motion (optical flow), as well as language cues (VLM). Each modality represents an axis of a polygon, streamlined to add salient cues to RGB. PI-VAD includes two plug-in modules, namely Pseudo-modality Generation module and Cross Modal Induction module, which generate modality-specific prototypical representation and, thereby, induce multi-modal information into RGB cues. These modules operate by performing anomaly-aware auxiliary tasks and necessitate five modality backbones -- only during training. Notably, PI-VAD achieves state-of-the-art accuracy on three prominent VAD datasets encompassing real-world scenarios, without requiring the computational overhead of five modality backbones at inference.
title Just Dance with $π$! A Poly-modal Inductor for Weakly-supervised Video Anomaly Detection
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2505.13123