Abstract
With the massive growth of events on the Internet, efficient organization and monitoring of events becomes a practical challenge. To deal with this problem, we propose a novel CO-PMHT (CO-Probabilistic Multi-Hypothesis Tracking) algorithm for cross-domain multi-event tracking to obtain their informative summary details and evolutionary trends over time. We collect a large-scale dataset by searching keywords on two domains (Gooogle News and Flickr) and downloading both images and textual content for an event. Given the input data, our algorithm can track multiple events in the two domains collaboratively and boost the tracking performance. Specifically, the bridge between two domains is a semantic posterior probability, that avoids the domain gap. After tracking, we can visualize the whole evolutionary process of the event over time and mine the semantic topics of each event for deep understanding and event prediction. The extensive experimental evaluations on the collected dataset well demonstrate the effectiveness of the proposed algorithm for cross-domain multi-event tracking.
- J. Allan. 2002. Topic Detection and Tracking: Event Based Information Retrieval. Kluwer Academic Press. Google Scholar
Digital Library
- Y. Bar-Shalom, F. Daum, and J. Huang. 2009. The probabilistic data association filter. IEEE Control Syst. Mag. 29, 6, 82--100.Google Scholar
Cross Ref
- D. M. Blei, A. Y. Ng, and M. I. Jordan. 2003. Latent dirichlet allocation. J. Mach. Learn. Res. 3, 3--4, 993--1022. Google Scholar
Digital Library
- H. L. Chieu and Y. K. Lee. 2004. Query based event extraction along a timeline. In Proceedings of the 27th Annual ACM SIGIR International Conference on Research and Development in Information Retrieval. Google Scholar
Digital Library
- I. Cox and S. Hingorani. 1979. An efficient implementation of reid's multiple hypothesis tracking algorithm and its evaluation for the purpose of visual tracking. IEEE Trans. Pattern Anal. Mach. Intell. 24, 6, 843--854.Google Scholar
- A. Dempster, N. Laird, and D. Rubin. 1977. Maximum likelihood from incomplete data via the em algorithm. J. Royal Statist. Soc. B39, 1, 1--38.Google Scholar
- P. Duygulu, J. Y. Pan, and D. A. Forsyth. 2005. Towards auto-documentary: Tracking the evolution of news stories. In Proceedings of the 12th Annual ACM International Conference on Multimedia (Multimedia'05). 820--827. Google Scholar
Digital Library
- J. R. Kender and M. R. Naphade. 2005. Visual concepts for news story tracking: Analyzing and exploiting the nist trecvid video annotation experiment. In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05). 1174--1181. Google Scholar
Digital Library
- Z. Knan, T. Balch, and F. Dellaert. 2005. MCMC-based particle filtering for tracking a variable number of interacting targets. IEEE Trans. Pattern Anal. Mach. Intell. 27, 11, 82--100. Google Scholar
Digital Library
- S. Kullback and R. A. Leibler. 1951. On information and sufficiency. Ann. Math. Statist. 22, 1, 79--86.Google Scholar
Cross Ref
- R. Kumar, U. Mahadevan, and D. Sivakumar. 2004. A graph-theoretic approach to extract storylines from search results. In Proceedings of the 10th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Google Scholar
Digital Library
- F. Lin and C.-H. Liang. 2008. Storyline-based summarization for news topic retrospection. Decis. Support Syst. 45, 4, 473--490. Google Scholar
Digital Library
- L. Liu, L. Wang, and X. Liu. 2011. In defense of soft assignment coding. In Proceedings of the IEEE International Conference on Computer Vision (ICCV'11). 2486--2493. Google Scholar
Digital Library
- S. Liu, S. Yan, T. Zhang, C. Xu, J. Liu, and H. Lu. 2012. Weakly-supervised graph propagation towards collective image parsing. IEEE Trans. Multimedia 14, 2, 361--373. Google Scholar
Digital Library
- J. Makkonen, H. A. Myka, and M. Salmenkivi. 2004. Simple semantics in topic detection and tracking. Inf. Retr. 7, 3--4, 347--368. Google Scholar
Digital Library
- G. Salton and C. Buckley. 1988. Term-weighting approaches in automatic text retrieval. Inf. Process. Manag. 24, 5, 513--523. Google Scholar
Digital Library
- G. Salton, A. Wong, and C. S. Yang. 1975. A vector space model for automatic indexing. Comm. ACM 18, 1, 613--620. Google Scholar
Digital Library
- H. Shitrit, J. Berclaz, F. Fleuret, and P. Fua. 2011. Tracking multiple people under global appearance constraints. In Proceedings of the IEEE International Conference on Computer Vision (ICCV'11). 1--8. Google Scholar
Digital Library
- R. L. Streit and T. E. Luginbuhl. 1995. Probabilistic multi-hypothesis tracking. Tech. rep. NUWC-NPT 10, 428, Naval Undersea Warfare Center, Newport, RI.Google Scholar
- X. Wu, C.-W. Ngo, and A. G. Hauptmann. 2008. Multimodal news story clustering with pairwise visual near-duplicate constraint. IEEE Trans. Multimedia 10, 4, 188--199. Google Scholar
Digital Library
- L. Xie, L. S. Kennedy, S.-F. Chang, A. Divakaran, H. Sun, and C.-Y. Lin. 2004. Discovering meaningful multimedia patterns with audio-visual concepts and associated text. In Proceedings of the International Conference on Image Processing (ICIP'04). 2383--2386.Google Scholar
Cross Ref
- J. Xing, H. Ai, and S. Lao. 2009. Multi-object tracking through occlusions by local tracklets filtering and global tracklets association with detection responses. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR'09). 1--8.Google Scholar
- Y. Yang, J. Carbonell, R. Brown, T. Pierce, B. Archibald, and X. Liu. 1999. Learning approaches for detecting and tracking news events. IEEE Intell. Syst. Appl. Intell. Inf. Retr. 14, 4, 32--43. Google Scholar
Digital Library
- Q. Yu and G. Medioni. 2009. Multiple-target tracking by spatio-temporal monte carlo markov chain data association. IEEE Trans. Pattern Anal. Mach. Intell. 31, 12, 1--110. Google Scholar
Digital Library
- Y. Zhai and M. Shah. 2005. Tracking news stories across different sources. In Proceedings of the 13th Annual ACM International Conference on Multimedia (Multimedia'05). 2--10. Google Scholar
Digital Library
- T. Zhang, B. Ghanem, S. Liu, and N. Ahuja. 2012a. Low-rank sparse learning for robust visual tracking. In Proceedings of the 12th European Conference on Computer Vision (ECCV'12). Google Scholar
Digital Library
- T. Zhang, C. Xu, G. Zhu, S. Liu, and H. Lu. 2012b. A generic framework for video annotation via semi-supervised learning. IEEE Trans. Multimedia 14, 4, 1206--1219. Google Scholar
Digital Library
- T. Zhang, B. Ghanem, S. Liu, and N. Ahuja. 2012c. Robust visual tracking via multi-task sparse learning. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR'12). 2042--2049. Google Scholar
Digital Library
- T. Zhang, B. Ghanem, S. Liu, and N. Ahuja. 2013a. Robust visual tracking via structured multi-task sparse learning. Int. J. Comput. Vis. 101, 2, 367--383. Google Scholar
Digital Library
- T. Zhang, B. Ghanem, S. Liu, C. Xu, and N. Ahuja. 2013b. Low-rank sparse coding for image classification. In Proceedings of the IEEE International Conference on Computer Vision (ICCV'13). Google Scholar
Digital Library
- T. Zhang, S. Liu, C. Xu, and H. Lu. 2013c. Mining semantic context information for intelligent video surveillance of traffic scenes. IEEE Trans. Industr. Informat. 9, 1, 149--160.Google Scholar
Cross Ref
Index Terms
Cross-Domain Multi-Event Tracking via CO-PMHT
Recommendations
Cross-view action recognition by cross-domain learning
This paper proposes a novel cross-view human action recognition method by discovering and sharing common knowledge among different video sets captured in multiple viewpoints. We treat a specific view as target domain and the others as source domains and ...
Dynamic video mix-up for cross-domain action recognition
Highlights- A video-level mix-up learning method for cross-domain action recognition.
- The ...
AbstractIn recent years, action recognition has been extensively studied. For some general action datasets, such as UCF101 [1], the recognition accuracy in a specific domain can reach 95%. However, due to the existence of the domain-wise ...
Cross-Domain Collaborative Learning in Social Multimedia
MM '15: Proceedings of the 23rd ACM international conference on MultimediaCross-domain data analysis is one of the most important tasks in social multimedia. It has a wide range of real-world applications, including cross-platform event analysis, cross-domain multi-event tracking, cross-domain video recommendation, etc. It is ...






Comments