.png){kind=icon}
.png){kind=icon}
Return
Multiple instance-based video anomaly detection using deep temporal encoding-decoding
.png){kind=icon}
OA
AI
DOI：10.1016/j.eswa.2022.119079.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In this paper, we propose a weakly supervised deep temporal encoding-decoding solution for anomaly detection in surveillance videos using multiple instance learning. The proposed approach uses both abnormal and normal video clips during the training phase which is developed in the multiple instance learning framework where we treat the video as a bag and video clips as instances in the bag. Our main contribution lies in the proposed novel approach to consider temporal relations between video instances in a weakly supervised setting. We deal with video instances (clips) as sequential visual data rather than a set of independent instances. We employ a deep temporal encoding-decoding network that is designed to capture spatio-temporal evolution of video instances over time. We also propose a new loss function that maximizes the mean distance between normal and abnormal instance predictions. The new loss function ensures a low false alarm rate which is crucial in practical surveillance applications. The proposed temporal encoding-decoding approach with the modified loss is benchmarked against the state of the art. The results show that the proposed method performs similar to, or better than the state-of-the-art solutions for anomaly detection in video surveillance applications and achieves the lowest false alarm rate on UCF-Crime dataset.
Keywords:
Anomaly detection
Surveillance videos
Weakly supervised multiple instance learning
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
7.5
Papers:
2.9W
Citations:
10.2W
Organization
No organization information available