Saved in:
Bibliographic Details
Main Authors: Pillai, Gargi V., Verma, Ashish, Sen, Debashis
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.00670
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917942757687296
author Pillai, Gargi V.
Verma, Ashish
Sen, Debashis
author_facet Pillai, Gargi V.
Verma, Ashish
Sen, Debashis
contents Anomaly detection in videos is a challenging task as anomalies in different videos are of different kinds. Therefore, a promising way to approach video anomaly detection is by learning the non-anomalous nature of the video at hand. To this end, we propose a one-class few-shot learning driven transformer based approach for anomaly detection in videos that is self-context aware. Features from the first few consecutive non-anomalous frames in a video are used to train the transformer in predicting the non-anomalous feature of the subsequent frame. This takes place under the attention of a self-context learned from the input features themselves. After the learning, given a few previous frames, the video-specific transformer is used to infer if a frame is anomalous or not by comparing the feature predicted by it with the actual. The effectiveness of the proposed method with respect to the state-of-the-art is demonstrated through qualitative and quantitative results on different standard datasets. We also study the positive effect of the self-context used in our approach.
format Preprint
id arxiv_https___arxiv_org_abs_2503_00670
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Transformer Based Self-Context Aware Prediction for Few-Shot Anomaly Detection in Videos
Pillai, Gargi V.
Verma, Ashish
Sen, Debashis
Computer Vision and Pattern Recognition
Anomaly detection in videos is a challenging task as anomalies in different videos are of different kinds. Therefore, a promising way to approach video anomaly detection is by learning the non-anomalous nature of the video at hand. To this end, we propose a one-class few-shot learning driven transformer based approach for anomaly detection in videos that is self-context aware. Features from the first few consecutive non-anomalous frames in a video are used to train the transformer in predicting the non-anomalous feature of the subsequent frame. This takes place under the attention of a self-context learned from the input features themselves. After the learning, given a few previous frames, the video-specific transformer is used to infer if a frame is anomalous or not by comparing the feature predicted by it with the actual. The effectiveness of the proposed method with respect to the state-of-the-art is demonstrated through qualitative and quantitative results on different standard datasets. We also study the positive effect of the self-context used in our approach.
title Transformer Based Self-Context Aware Prediction for Few-Shot Anomaly Detection in Videos
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.00670