Analyzing multiple types of behaviors from traffic videos via nonparametric topic model

Abstract Intelligent video surveillance systems have garnered substantial research attention in recent years within the transportation surveillance field. The systems can assist in identifying activities, interactions, and abnormal behaviors of individuals in traffic. We propose a novel unsupervised learning framework based on a two-layer BNBP-PFA topic model to simultaneously model multiple types of behaviors in crowded and complicated traffic videos. We provide the model’s structure, its inference algorithm, and design a corresponding likelihood function based on an abnormality detection algorithm. Compared to similar existing algorithms, ours readily reveals both the local topic-motion pattern and the global topic-traffic pattern. Comparative experiments on two public traffic video datasets show that our model outperforms the state-of-art algorithms in regards to effective topic discovery and abnormality detection.

[1]  Shaogang Gong,et al.  A Markov Clustering Topic Model for mining behaviour in video , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[2]  Michael I. Jordan,et al.  Latent Dirichlet Allocation , 2001, J. Mach. Learn. Res..

[3]  Tieniu Tan,et al.  A system for learning statistical motion patterns , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  Michael I. Jordan,et al.  Hierarchical Dirichlet Processes , 2006 .

[5]  Hanqing Lu,et al.  Learning Semantic Motion Patterns for Dynamic Scenes by Improved Sparse Topical Coding , 2012, 2012 IEEE International Conference on Multimedia and Expo.

[6]  David B. Dunson,et al.  Beta-Negative Binomial Process and Poisson Factor Analysis , 2011, AISTATS.

[7]  Eric P. Xing,et al.  Sparse Topical Coding , 2011, UAI.

[8]  Tao Xiang,et al.  Identifying Rare and Subtle Behaviors: A Weakly Supervised Joint Topic Model , 2011, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[9]  Iman Gholampour,et al.  Modeling traffic motion patterns via Non-negative Matrix Factorization , 2015, 2015 IEEE International Conference on Signal and Image Processing Applications (ICSIPA).

[10]  Iman Gholampour,et al.  Incorporating fully sparse topic models for abnormality detection in traffic videos , 2014, 2014 4th International Conference on Computer and Knowledge Engineering (ICCKE).

[11]  Xueqi Cheng,et al.  Group sparse topical coding: from code to topic , 2013, WSDM.

[12]  Hua Yang,et al.  Geometric Motion Flow (GMF): A New Feature for Traffic Surveillance , 2011, 2011 Sixth International Conference on Image and Graphics.

[13]  P. Paatero,et al.  Positive matrix factorization: A non-negative factor model with optimal utilization of error estimates of data values† , 1994 .

[14]  Chong Wang,et al.  The IBP Compound Dirichlet Process and its Application to Focused Topic Modeling , 2010, ICML.

[15]  Tu Bao Ho,et al.  Fully Sparse Topic Models , 2012, ECML/PKDD.

[16]  Zhongke Shi,et al.  Understanding dynamic scenes by hierarchical motion pattern mining , 2011, 2011 IEEE International Conference on Multimedia and Expo.

[17]  Mubarak Shah,et al.  Learning object motion patterns for anomaly detection and improved object detection , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[18]  Iman Gholampour,et al.  Discovering motion patterns in traffic videos using improved Group Sparse Topical Coding , 2014, 7'th International Symposium on Telecommunications (IST'2014).

[19]  W. Eric L. Grimson,et al.  Unsupervised Activity Perception by Hierarchical Bayesian Models , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.

[20]  Lawrence Carin,et al.  Negative Binomial Process Count and Mixture Modeling , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[21]  Bodo Rosenhahn,et al.  Video Event Recognition by Combining HDP and Gaussian Process , 2015, 2015 IEEE International Conference on Computer Vision Workshop (ICCVW).