Object-Centric Auto-Encoders and Dummy Anomalies for Abnormal Event Detection in Video

By Radu Tudor Ionescu, Fahad Shahbaz Khan, Mariana-Iuliana Georgescu, Ling Shao. Computer Vision and Pattern Recognition, 2018

395 citations recorded by the source corpus.

Abnormal event detection in video is a challenging vision problem. Most existing approaches formulate abnormal event detection as an outlier detection task, due to the scarcity of anomalous data during training. Because of the lack of prior information regarding abnormal events, these methods are not fully-equipped to differentiate between normal and abnormal events. In this work, we formalize abnormal event detection as a one-versus-rest binary classification problem. Our contribution is two-fold. First, we introduce an unsupervised feature learning framework based on object-centric convolutional auto-encoders to encode both motion and appearance information. Second, we propose a supervised classification approach based on clustering the training samples into normality clusters. A one-versus-rest abnormal event classifier is then employed to separate each normality cluster from the rest. For the purpose of training the classifier, the other clusters act as dummy anomalies. During inference, an object is labeled as abnormal if the highest classification score assigned by the one-versus-rest classifiers is negative. Comprehensive experiments are performed on four benchmarks: Avenue, ShanghaiTech, UCSD and UMN. Our approach provides superior results on all four data sets. On the large-scale ShanghaiTech data set, our method provides an absolute gain of 8.4% in terms of frame-level AUC compared to the state-of-the-art method.

Subfields: Artificial Intelligence, Computer Networks and Communications, Epidemiology

View Paper: https://www.semanticscholar.org/paper/e7b7d97042ad2fdf3a7238a724c9dc3195537bea

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Fahad Shahbaz Khan
2. Chunhua Shen
3. Anton van den Hengel
4. Guansong Pang
5. Mubarak Shah
6. Radu Tudor Ionescu
7. Mariana-Iuliana Georgescu
8. Longbing Cao

In Computer Networks and Communications

1. Guansong Pang
2. Chunhua Shen
3. Radu Tudor Ionescu
4. Fahad Shahbaz Khan
5. Mariana-Iuliana Georgescu
6. Anton van den Hengel
7. Longbing Cao
8. Gustavo Carneiro

In Epidemiology

1. Fahad Shahbaz Khan
2. Guansong Pang
3. Mubarak Shah
4. Mariana-Iuliana Georgescu
5. Radu Tudor Ionescu
6. Anton van den Hengel
7. Chunhua Shen
8. Longbing Cao

Citations in the corpus

• Deep Learning for Anomaly Detection (2020, methodology): “OADA [65] (4) Semi Reconstruction Yes No No AE ReLU 3 MSE Video”
• Self-Supervised Predictive Convolutional Attentive Block for Anomaly Detection (2021, methodology): “Anomaly detection methods can be classified into: dictionary learning methods [7–9, 14, 40, 55], change detection frameworks [11, 26, 38, 48], probability-based methods [1, 2, 16, 23, 29, 44, 45, 56, 61, 74], distance-based models [6, 10, 25, 27, 50, 51,53,57,58,63,65,68,70], and reconstruction-base”
• Weakly-supervised Video Anomaly Detection with Robust Temporal Feature Magnitude Learning (2021, background): “Al-ternatively, some approaches depend on data reconstruction using generative models to learn the representations of normal samples by (adversarially) minimising the reconstruction error [6,14,18,18,28,35–37,42,49,52,53,62,68,82].”
• Anomaly Detection in Video via Self-Supervised and Multi-Task Learning (2020, methodology): “Unlike [9, 10, 17, 61], we perform the analysis through a series of proxy self-supervised and model distillation tasks, proposing a novel anomaly detection framework based on multi-task learning.”