Janus - Multi Source Event Detection and Collection System for Effective Surveillance of Criminal Activity
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Recent technological advances provide the opportunity to use large amountsof multimedia data from a multitude of sensors with different modalities (e.g., video, text)for the detection and characterization of criminal activity. Their integration cancompensate for sensor and modality deficiencies by using data from other availablesensors and modalities. However, building such an integrated system at the scale ofneighborhood and cities is challenging due to the large amount of data to be consideredand the need to ensure a short response time to potential criminal activity. In this paper, we present a system that enables multi-modal data collection at scaleand automates the detection of events of interest for the surveillance and reconnaissanceof criminal activity. The proposed system showcases novel analytical tools that fusemultimedia data streams to automatically detect and identify specific criminal events andactivities. More specifically, the system detects and analyzes series of incidents (anincident is an occurrence or artifact relevant to a criminal activity extracted from a singlemedia stream) in the spatiotemporal domain to extract events (actual instances ofcriminal events) while cross-referencing multimodal media streams and incidents in timeand space to provide a comprehensive view to a human operator while avoidinginformation overload. We present several case studies that demonstrate how theproposed system can provide law enforcement personnel with forensic and real time toolsto identify and track potential criminal activity.
Abstract
Recent technological advances provide the opportunity to use large amountsof multimedia data from a multitude of sensors with different modalities (e.g., video, text)for the detection and characterization of criminal activity. Their integration cancompensate for sensor and modality deficiencies by using data from other availablesensors and modalities. However, building such an integrated system at the scale ofneighborhood and cities is challenging due to the large amount of data to be consideredand the need to ensure a short response time to potential criminal activity. In this paper, we present a system that enables multi-modal data collection at scaleand automates the detection of events of interest for the surveillance and reconnaissanceof criminal activity. The proposed system showcases novel analytical tools that fusemultimedia data streams to automatically detect and identify specific criminal events andactivities. More specifically, the system detects and analyzes series of incidents (anincident is an occurrence or artifact relevant to a criminal activity extracted from a singlemedia stream) in the spatiotemporal domain to extract events (actual instances ofcriminal events) while cross-referencing multimodal media streams and incidents in timeand space to provide a comprehensive view to a human operator while avoidinginformation overload. We present several case studies that demonstrate how theproposed system can provide law enforcement personnel with forensic and real time toolsto identify and track potential criminal activity.
참고문헌 (26)
- W. Zajdel, "CASSANDRA : audio-video sensor fusion for aggression detection", Proceedings of IEEE Conference on Advanced Video and Signal Based Surveillance, pp.200~205 (2007)
- C. G. M. Snoek, "Multimodal Video Indexing : A Review of the State-of-the-art", Multimedia Tools and Applications, 25, pp.5~35 (2005)
- J. Kang, "Automatic Tracking and Labeling of Human Activities in a Video Sequence", Proceedings of the 6th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance (2004)
- J. Kang, "Persistent Objects Tracking Across Multiple Non Overlapping Cameras", IEEE Workshop on Motion and Video Computing (2005)
- L. Yaroslavsky, "Processing and Fusion of Thermal and Video Sequences for Terrestrial Long Range Observation Systems", Proceedings the 7th International Conference on Information Fusion, pp.848~855 (2004)
- B. Fishbain, "Spatial, Temporal, and Inter-channel Image Data Fusion for Long-Distance Terrestrial Observation Systems", Advances in Optical Technologies, 2008 (2008)
- M. Valera, "Intelligent distributed surveillance systems: a review", IEEE Proceedings of Vision, Image and Signal Processing, 152, pp.192~204 (2005)
- K. Toyama, "Geographic Location Tags on Digital Images", Proceedings of the ACM International Conference on Multimedia, pp.156~166 (2003)
- A. Pigeau, "Building and Tracking Hierarchical Geographical & Temporal Partitions for Image Collection Management on Mobile Devices", Proceedings of the ACM International Conference on Multimedia (2005)
- I. Simon, "Scene Segmentation Using the Wisdom of Crowds", Proceedings of European Conference on Computer Vision (2008)
- Y. Li, "Location Recognition using Prioritized Feature Matching", Proceedings of European Conference on Computer Vision (2010)
- S. Agarwal, "Building Rome in a Day", Proceedings of International Conference on Computer Vision (2009)
- T. Zhao, "Segmentation and Tracking of Multiple Humans in Crowded Environments", IEEE Transactions on Pattern Analysis and Machine Intelligence, 30(7), pp.1198~1211 (2008)
- B. Wu, "Detection and Tracking of Multiple, Partially Occluded Humans by Bayesian Combination of Edgelet based Part Detectors", International Journal of Computer Vision, 75, pp.247~266 (2007)
- G. Medioni, "Event detection and analysis from video streams", IEEE Transactions on Pattern Analysis and Machine Intelligence, 23, pp.873~889 (2001)
- S.A. Ay, "Viewable Scene Modeling for Geospatial Video Search", Proceedings of the ACM International Conference on Multimedia, pp.309~318 (2008)
- S.A. Ay, "GRVS: a georeferenced video search engine", ACM Multimedia, ~ (2009)
- S. Kim, "Design and Implementation of Geo-Tagged Video Search Framework", Journal of Visual Communications and Image Representation Special Issue on Large Scale Image and Video Search; Challenges, Technologies, and Trends, 21(8), pp.773~786 (2010)
- Y. Cai, "Towards a practical ptz face detection and tracking system", WACV (2013)
- C. Huang, "High performance object detection by collaborative learning of joint ranking of granules features", CVPR, pp.41~48 (2010)
- B. Yang, "Online learned discriminative part-based appearance models for multihuman tracking", ECCV (2012)
- C. Huang, "Robust Object Tracking by Hierarchical Association of Detection Responses", ECCV (2008)
- Ali Khodaei, "Hybrid Indexing and Seamless Ranking of Spatial and Textual Features of Web Documents", 21st International Conference on Database and Expert Systems Applications (DEXA10) (2010)
- Ali Khodaei, "SKIF-P: a point-based indexing and ranking of web documents for spatial-keyword search", Geoinformatica (2011), Springer Netherlands[DOI]
- Ali Khodaei, "Temporal-Textual Retrieval: Time and Keyword Search in Web Documents", International Journal of Next-Generation Computing (2012)
- Chi Y., "Catch the Moment: Maintaining Closed Frequent Itemsets in a Data Stream Sliding Window", Knowledge and Information Systems, 10(3), pp.265~294 (2006)