Anomaly Detection Results

Anomaly Detection Results

This page shows results found in the literature for anomalous event detection in crowd data sets. If you like to have your published results added in the following tables, please send an e-mail to Rensso Mora with the link (or the pdf) to your paper and the results to be reported. Up to now, we have tabulated the results for the following datasets.

  1. University of California, San Diego (UCSD), this data set contains two sequences Peds1 and Peds2
  2. UMN
  3. Subway Entrance-Exit

 

UCSD Dataset

Method AUC Peds1 Error Peds1 AUC Peds2 Error Peds2 year
HOMF [1] 0.71 33.6 0.89 19.0 2015
MDT-Temp [2] 0.825 22.9 0.765 27.9 2014
MPPCA [3] 0.674 35.6 0.71 35.8 2009
Force Flow [4] 0.688 36.5 0.702 35.0 2009
GPR [5] 0.838 23.7 2015
OADC [6] 0.91 0.925 2015
K-NN [7] 0.927 16.0 2012
AMDN [8] 0.921 16.0 0.90 17.0 2015
CNN-Anom [9] 0.87 24.0 0.88 24.4 2016
STT [10] 0.87 20.0 2015
F-Matlab [11] 0.918 15.0 2013
Social-Attribute[12] 0.782 26.0 2015
Gaussian-KNN[13] 0.927 16.0 2015
Sparce-code[14] 0.9243 15.0 2014
Scan Statistic[15] 0.87 0.94 2013
Video Parsing[16] 0.91 18.0 0.92 14.0 2011
2D-HMM[17] 0.859 21.68 0.928 11.67 2012
HNC[18] 10.0 10.0 2015
UCSD (Frame Level)

[1] Rensso Victor Hugo Mora Colque, Carlos Caetano, William Robson Schwartz, “Histograms of Optical Flow Orientation and Magnitude to Detect Anomalous Events in Videos,” in SIBGRAPI, 2015.

[2] W. Li, V. Mahadevan, N. Vasconcelos, “Anomaly detection and localization in crowded scenes,” in IEEE Transactions on, 2014.

[3] J. Kim, K. Grauman , “Observe locally, infer globally: A spacetime mrf for detecting abnormal activities with incremental updates,” in CVPR, 2009.

[4] R. Mehran, A. Oyama, M. Shah, “Abnormal crowd behavior detection using social force model,” in CVPR, 2009.

[5] Kai-Wen Cheng, Yie-Tarng Chen, Wen-Hsien Fang “Video Anomaly Detection and Localization Using Hierarchical Feature Representation and Gaussian Process Regression,” in CVPR, 2015.

[6] Y. Yuan and J. Fang and Q. Wang “Online Anomaly Detection in Crowd Scenes via Structure Analysis,” in IEEE Transactions On Cybernetics, 2015.

[7] V. Saligrama, Z. Chen “Video anomaly detection based on local statistical aggregates,” in CVPR, 2012.

[8] Xu, D., Ricci, E., Yan, Y., Song, J., Sebe, N., Kessler, F.B. “Learning deep representations of appearance and motion for anomalous event detection ,” in BMVC, 2015.

[9] Zhou, S., Shen, W., Zeng, D., Fang, M., Wei, Y., & Zhang, Z. “Spatial-temporal convolutional neural networks for anomaly detection and localization in crowded scenes ,” in Signal Processing: Image Communication, 2016.

[10] Zhang, Y., Lu, H., Zhang, L., Ruan, X., & Sakai, S.“Video anomaly detection based on locality sensitive hashing filters,” in Pattern Recognition, 2015.

[11] Lu, C., Shi, J., & Jia, J. “Abnormal Event Detection at 150 FPS in MATLAB,” in International Conference on Computer Vision, 2013.

[12] Zhang, Y., Qin, L., Ji, R., Yao, H., & Huang, Q. “Social Attribute-Aware Force Model: Exploiting Richness of Interaction for Abnormal Crowd Detection,” in IEEE Transactions on Circuits and Systems for Video Technology, 2015.

[13] Cheng, K., Chen, Y., & Fang, W. “ Gaussian Process Regression-Based Video Anomaly Detection and Localization With Hierarchical Feature Representation,” in IEEE Transactions on Image Processing, 2015.

[14] Liu, Y., Li, Y., & Ji, X. “ Abnormal Event Detection in Nature Settings,” in International Journal of Signal Processing, Image Processing and Pattern Recognition, 2014.

[15] Hu, Y., Zhang, Y., & Davis, L. S. “ Unsupervised abnormal crowd activity detection using semiparametric scan statistic,” in IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, 2013.

[16] Antić, B., & Ommer, B. “ Video parsing for abnormality detection,” in Proceedings of the IEEE International Conference on Computer Vision, 2011.

[17] Nallaivarothayan, H., Ryan, D., Denman, S., Sridharan, S., & Fookes, C. “ Anomalous Event Detection Using a Semi-Two Dimensional Hidden Markov Model ,” in In Digital Image Computing Techniques and Applications (DICTA), 2012.

[18] Xiao, T., Zhang, C., & Zha, H. “Learning to detect anomalies in surveillance video ,” in IEEE Signal Processing Letters, 2015.

UMN Dataset

Method AUC ERR year
Social Force[1] 0.96 2009
K-NN[2] 0.98 2012
Multi_scale motion[3] 0.996 2013
Scan statistic[4] 0.991 2013
Motion influence[5] 0.995 2013
OACD[6] 0.9887 2014
MDT-Temp[7] 0.996 2.8 2014
Sparce code[8] 0.992 2014
OADC-SA[9] 0.9967 2015
Social attribute[10] 0.986 2015
STT[11] 0.983 2015
CNN-anom[12] 0.9963 2015
UMN dataset – top ranked results (in %)

[1] Mehran, R., Oyama, A., & Shah, M. “ Abnormal crowd behavior detection using social force model,” in IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, 2009.

[2] V. Saligrama, Z. Chen “Video anomaly detection based on local statistical aggregates,” in CVPR, 2012.

[3] Du, D., Qi, H., Huang, Q., Zeng, W., & Zhang, C. “ Abnormal event detection in crowded scenes based on Structural Multi-scale Motion Interrelated Patterns,” in Proceedings – IEEE International Conference on Multimedia and Expo, 2013.

[4] Hu, Y., Zhang, Y., & Davis, L. S. “ Unsupervised abnormal crowd activity detection using semiparametric scan statistic,” in IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, 2013.

[5] Lee, D. G., Suk, H. Il, & Lee, S. W. “ Crowd behavior representation using motion influence matrix for anomaly detection,” in 2nd IAPR Asian Conference on Pattern Recognition, 2013.

[6] Liu, Y., Li, Y., & Ji, X. “ Abnormal Event Detection in Nature Settings,” in International Journal of Signal Processing, Image Processing and Pattern Recognition, 2014.

[7] W. Li, V. Mahadevan, N. Vasconcelos, “Anomaly detection and localization in crowded scenes,” in IEEE Transactions on, 2014.

[8] Liu, Y., Li, Y., & Ji, X. “ Abnormal Event Detection in Nature Settings,” in International Journal of Signal Processing, Image Processing and Pattern Recognition, 2014.

[9] Yuan, Y., Fang, J., & Wang, Q. “Online anomaly detection in crowd scenes via structure analysis ,” in IEEE Transactions on Cybernetics, 2015.

[10] Zhang, Y., Qin, L., Ji, R., Yao, H., & Huang, Q. “Social Attribute-Aware Force Model: Exploiting Richness of Interaction for Abnormal Crowd Detection.,” in IEEE Transactions on Circuits and Systems for Video Technology, 2015.

[11] Zhang, Y., Lu, H., Zhang, L., Ruan, X., & Sakai, S.“Video anomaly detection based on locality sensitive hashing filters,” in Pattern Recognition, 2015.

[12] Zhou, S., Shen, W., Zeng, D., Fang, M., Wei, Y., & Zhang, Z. “Spatial-temporal convolutional neural networks for anomaly detection and localization in crowded scenes ,” in Signal Processing: Image Communication, 2016.

Subway Entrance-Exit Dataset

Method AUC ERR year
MDT-temp*[1] 0.897-0.908 16.4-16.7 2014
GPR[2] 0.927 10.9 2015
CNN-Anom*[2] 0.927-0.919 2015
Subway entrance-exit dataset – top ranked results (in %)

(*) Some results are presented for entrance and exit sequences respectively.

[1] W. Li, V. Mahadevan, N. Vasconcelos, “Anomaly detection and localization in crowded scenes,” in IEEE Transactions on, 2014.

[2] Kai-Wen Cheng, Yie-Tarng Chen, Wen-Hsien Fang “Video Anomaly Detection and Localization Using Hierarchical Feature Representation and Gaussian Process Regression,” in CVPR, 2015.

[3] Zhou, S., Shen, W., Zeng, D., Fang, M., Wei, Y., & Zhang, Z. “Spatial-temporal convolutional neural networks for anomaly detection and localization in crowded scenes ,” in Signal Processing: Image Communication, 2016.