Anomaly detection in extremely crowded scenes using spatio-temporal motion pattern models

By Louis Kratz, Ko Nishino. 2009 IEEE Conference on Computer Vision and Pattern Recognition, 2009

739 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/2e269d2ee60db6d09a514c4748e3fdf9202917f9

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

In Computer Vision and Pattern Recognition

Citations in the corpus

Discovering Motion Primitives for Unsupervised Grouping and One-Shot Learning of Human Actions, Gestures, and Expressions (2013, positive): “The idea for the proposed representation is inspired by several recent works in traffic patterns discovery and representation [45], [20], [34].”
Learning Deep Representations of Appearance and Motion for Anomalous Event Detection (2015, neutral): “[12] introduce a HMMs-based approach for detecting abnormal events by analyzing the motion variation of local space-time volumes.”
Real-World Anomaly Detection in Surveillance Videos (2018, methodology): “Due to difficulties in obtaining reliable tracks, several approaches avoid tracking and learn global motion patterns through histogram-based methods [10], topic modeling [20], motion patterns [32], social force models [29], mixtures of dynamic textures model [27], Hidden Markov Model (HMM) on local ”
Graph Convolutional Label Noise Cleaner: Train a Plug-And-Play Action Classifier for Anomaly Detection (2019, background): “…via a variety of statistic models, e.g . the social force model [45], the mixture of dynamic models on texture [35], Hidden Markov Models on video volumes [21, 30], the Markov Random Field upon spatial-temporal domain [28], Gaussian process modeling [49, 11], and identify anomalies as outliers.”
Abnormal Event Detection at 150 FPS in MATLAB (2013, methodology): “We compute 3D gradient features on each of them following [11].”
On pairwise costs for network flow multi-object tracking (2014, methodology): “In the following subsections, we show how some traditional constraints [15, 21] could be incorporated in our”
Crowded Scene Analysis: A Survey (2015, methodology): “The proposed detector is based on a video representation that accounts for both , Kratz et al. [60] have presented a framework extremely crowded scenes.”
Crowd Analysis Using Computer Vision Techniques (2010, background): “Despite the good quantitative results reported in [45], the determination of the size of the cuboids can be challenging, particularly in the far field of oblique camera setups (due to perspective issues).”
Deep-anomaly: Fully convolutional neural network for fast anomaly detection in crowded scenes (2016, neutral): “In [37], a Gaussian model is fitted into spatio-temporal gradient features, and a hidden Markov model (HMM) is used for detecting the abnormal events.”
Spatio-Temporal AutoEncoder for Video Anomaly Detection (2017, methodology): “[8] presented a HMM-based approach to detect anomalies using 3D Gaussian distributions of spatio-temporal gradients.”
Weakly-supervised Video Anomaly Detection with Robust Temporal Feature Magnitude Learning (2021, methodology): “In anomaly detection, traditional methods [1, 8] convert consecutive frames into handcrafted motion trajectories to capture the local consistency between neighbouring frames.”
Chaotic invariants of Lagrangian particle trajectories for anomaly detection in crowded scenes (2010, background): “Limiting our review of the literature to work directly related to analysis of crowded scenes, common features used for modeling crowded motions include optical flow [9, 10, 11, 12], gradient [13], spatiotemporal volume [14], etc.”
A Review on Video-Based Human Activity Recognition (2013, background): “Kratz and Nishino [135] propose the distributions of spatio-temporal gradients as motion features.”
Video-Based Abnormal Human Behavior Recognition—A Review (2012, methodology): “(see [16], [23], [71], [109], [110], [116]–[118]).”
Spatio-Temporal Data Mining (2017, methodology): “For example, in the application of detecting abnormal activities in crowded scenes from surveillance camera videos, anomaly detection approaches using models of normal activity have been developed in [Kratz and Nishino 2009; Li et al. 2014a].”
Abnormal event detection in crowded scenes using sparse representation (2013, methodology): “There are several statistics models, such as Gaussian model, Gaussian Mixture Model (GMM) or Mixture Principle Component Analysis (MPPCA) [16], Hidden Markov Model (HMM) [17], Markov Random Field (MRF) [3] or spatio-temporal MRF [16], Latent Dirichlet Allocation (LDA) [34].”
Detecting anomalous events in videos by learning deep representations of appearance and motion (2017, methodology): “[38] introduced an HMMs-based approach for detecting abnormal events through analyzing the motion variation of local space-time volumes.”