Anomaly detection in traffic surveillance with sparse topic model

Abstract

Most research on anomaly detection has focused on event that is different from its spatial-temporal neighboring events. It is still a significant challenge to detect anomalies that involve multiple normal events interacting in an unusual pattern. In this work, a novel unsupervised method based on sparse topic model was proposed to capture motion patterns and detect anomalies in traffic surveillance. scale-invariant feature transform (SIFT) flow was used to improve the dense trajectory in order to extract interest points and the corresponding descriptors with less interference. For the purpose of strengthening the relationship of interest points on the same trajectory, the fisher kernel method was applied to obtain the representation of trajectory which was quantized into visual word. Then the sparse topic model was proposed to explore the latent motion patterns and achieve a sparse representation for the video scene. Finally, two anomaly detection algorithms were compared based on video clip detection and visual word analysis respectively. Experiments were conducted on QMUL Junction dataset and AVSS dataset. The results demonstrated the superior efficiency of the proposed method.

Keywords

motion pattern / sparse topic model / SIFT flow / dense trajectory / fisher kernel

References

[1] LeyvaR, SanchezV, LiC. Video anomaly detection based on wake motion descriptors and perspective grids [C]. IEEE International Workshop on Information Forensics and Security, 2015209214

[2] PiciarelliC, ForestiG L. Surveillance-oriented event detection in video streams [J]. IEEE Intelligent Systems, 2010, 26(3): 32-41

[3] CongY, YuanJ, LiuJ. Sparse reconstruction cost for abnormal event detection [C]. IEEE Conference on Computer Vision and Pattern Recognition, 201134493456

[4] RenH, MoeslundT B. Abnormal event detection using local sparse representation [C]. IEEE International Conference on Advanced Video and Signal Based Surveillance, 2014125130

[5] XiaL-m, YangB-j, TuH-bin. Recognition of suspicious behavior using case-based reasoning [J]. Journal of Central South University, 2015, 22(1): 241-250

[6] RoshtkhariM J, LevineM D. Online dominant and anomalous behavior detection in videos [C]. Computer Vision and Pattern Recognition, 201326112618

[7] BiswasS, BabuR V. Sparse representation based anomaly detection with enhanced local dictionaries [C]. IEEE International Conference on Image Processing, 201555325536

[8] CuiX, LiuQ, GaoM, MetaxasD N. Abnormal detection using interaction energy potentials [C]. Computer Vision and Pattern Recognition, 201131613167

[9] BeraA, KimS, ManochaD. Realtime anomaly detection using trajectory-level crowd behavior learning [C]. IEEE Conference on Computer Vision and Pattern Recognition Workshops, 201612891296

[10] MehranR, OyamaA, ShahM. Abnormal crowd behavior detection using social force model [C]. Conference on Computer Vision and Pattern Recognition, 2009935942

[11] AhmadiP, TabandehM, GholampourI. Abnormal event detection and localisation in traffic videos based on group sparse topical coding [J]. Let Image Processing, 2015, 10(3): 235-246

[12] ChengK W, ChenY T, FangW H. Gaussian process regression-based video anomaly detection and localization with hierarchical feature representation [J]. IEEE Transactions on Image Processing, 2015, 24(12): 5288-5301

[13] MoX, MongaV, BalaR, FanZ G. Adaptive sparse representations for video anomaly detection [J]. IEEE Transactions on Circuits & Systems for Video Technology, 2014, 24(4): 631-645

[14] KavianiR, AhmadiP, GholampourI. Incorporating fully sparse topic models for abnormality detection in traffic videos [C]. International Econference on Computer and Knowledge Engineering, 2014586591

[15] EmonetR, VaradarajanJ, OdobezJ M. Extracting and locating temporal motifs in video scenes using a hierarchical non parametric bayesian model [C]. Computer Vision and Pattern Recognition, 201132333240

[16] VaradarajanJ, EmonetR, OdobezJ M. A sequential topic model for mining recurrent activities from long term video logs [J]. International Journal of Computer Vision, 2013, 103(1): 100-126

[17] YooY, YunK, YunS, HongJ H, JeongH, ChoiJ Y. Visual path prediction in complex scenes with crowded moving objects [C]. Computer Vision and Pattern Recognition, 201626682677

[18] FuW, WangJ, LuH, MaS D. Dynamic scene understanding by improved sparse topical coding [J]. Pattern Recognition, 2013, 46(7): 1841-1850

[19] JeongH, YooY, YiK M, ChoiJ Y. Two-stage online inference model for traffic pattern analysis and anomaly detection [J]. Machine Vision & Applications, 2014, 25(6): 1501-1517

[20] PathakD, SharangA, MukerjeeA. Anomaly localization in topic-based analysis of surveillance videos [C]. IEEE Winter Conference on Applications of Computer Vision, 2015389395

[21] WangH, KläserA, SchmidC, LiuC L. Dense trajectories and motion boundary descriptors for action recognition [J]. International Journal of Computer Vision, 2013, 103(1): 60-79

[22] ZengH, MaK K, WangC, CaiC H. SIFT-flow-based color correction for multi-view video [J]. Signal Processing Image Communication, 2015, 36(C): 53-62

[23] PerronninF, NchezJ, MensinkT. Improving the fisher kernel for large-scale image classification [C]. European Conference on Computer Vision, 2010143156

[24] PerronninF, DanceC, CsurkaG, BressanM. Adapted vocabularies for generic visual categorization [C]. European Conference on Computer Vision, 2006464475

[25] EltoftT, KimT, LeeT W. On the multivariate Laplace distribution [J]. IEEE Signal Processing Letters, 2006, 13(5): 300-303

[26] WangJ, FuW, LuH, MaS D. Bilayer sparse topic model for scene analysis in imbalanced surveillance videos [J]. IEEE Transactions on Image Processing, 2014, 23(12): 5198-5208