Anomaly detection in crowded scenes

By Vijay Mahadevan, Wei-Xin Li, V. Bhalodia, Nuno Vasconcelos. 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2010

1,581 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Biomedical Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/9d3f0d47449c7db37d1bae3b70db2928610a8db7

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 Biomedical Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Abnormal event detection in crowded scenes using sparse representation (2013, methodology): “Examples of abnormal detections using (i) the MDT approach [23], (ii) the SF-M S) and (iv) our approach with LSDS, where our approach with or without LSDS kground regions; and for SF-MPPCA, it completely misses the skater in (b), the perso on walking on the grass in (c).”
Abnormal Event Detection at 150 FPS in MATLAB (2013, methodology): “Our configuration is similar to that of [13].”
Spatio-Temporal AutoEncoder for Video Anomaly Detection (2017, methodology): “The optical flow input is not able to improve the results on UCSD Pedestrian dataset because the outlier in this dataset are rather appearance anomalies while the appearance features are not contained by the optical flow.”
Future Frame Prediction for Anomaly Detection - A New Baseline (2017, methodology): “Lots of efforts have been made for anomaly detection [20][13][23].”
Learning Temporal Regularity in Video Sequences (2016, methodology): “They are curated for anomaly or abnormal event detection and are referred to as Avenue [8], UCSD pedestrian [12], and Subway [11] datasets.”
A Review on Video-Based Human Activity Recognition (2013, methodology): “[132] construct probabilistic dynamic model using Mixtures of Dynamic Textures (MDT) [133].”
Anomaly Detection and Localization in Crowded Scenes (2014, methodology): “The initial value of w is learned, assuming a logistic regression model (vT 1⁄4 vS 1⁄4 0 in (16)), with the procedure of [43].”
Deep-anomaly: Fully convolutional neural network for fast anomaly detection in crowded scenes (2016, neutral): “Two measures at frame level and pixel level are used, which are introduced in [33] and often exploited in later work.”
Abnormal Event Detection in Videos using Spatiotemporal Autoencoder (2017, methodology): “We train our model on ﬁve most commonly used benchmarking datasets: Av-enue [13], UCSD Ped1 and Ped2 [14], Subway entrance and exit datasets [1].”
Deep Learning for Anomaly Detection (2020, background): “ults to define anomalies,e.g., cluster size [67], distance to cluster centers [60], distance between cluster centers [68], and cluster membership [142]. Gaussian mixture model-based anomaly detection [44, 96] is also included into this category due to some of its intrinsic relations to clustering, e”
Learning Regularity in Skeleton Trajectories for Anomaly Detection in Videos (2019, methodology): “It prevents us from trying our method on UCSD Ped1/Ped2 [21], another popular dataset whose video quality is too low to detect skeletons.”
Sparse reconstruction cost for abnormal event detection (2011, background): “Mahadevan et al.[18] model the normal crowd behavior by mixtures of dynamic textures.”
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”
Video anomaly detection based on local statistical aggregates (2012, methodology): “A key insight of the paper is that if anomalies are local optimal decision rules are local even when the nominal behavior exhibits global spatial and temporal statistical dependencies.”
Fighting Fake News: Image Splice Detection via Learned Self-Consistency (2018, methodology): “We trained our models, including the variations, using a ResNet50 [37] pretrained on ImageNet [50].”