Research on Various Methods for Visual Concept Detection

Abstract:

Visual concept detection is a functional way to detect, manage and classify visual information. No matter it is image retrieval or video retrieval, the system can distinguish object and scene concepts by learning image descriptions itself. However, when the viewing conditions is different or the viewing points is changed, all these will lead to a various result of the image description which collect by the system. So an effective visual concept classification methods should be invariant to different and accidental recording circumstances or conditions. This paper has analysis and summarize the current research status and detection methods of Visual concept detection technologies. Including such methods: salient point detection; social tagged images as a training resource for automated concept detection; extracting distinctive invariant features from images; cross-domain kernel learning method for visual concept detection, etc. And in the end, this paper seeks to unravel the effective when using MIR Flickr in visual concept detection. After comparasion, we want to find the advantages and disadvantages of these typical methods, and hope to give a valuable reference to the following researchers who is interested in visual concept detection and classifications.

• Periodical: Advanced Materials Research (Volumes 616-618)
• Pages: 2171-2174
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.616-618.2171
• Online since: December 2012
• Authors: Yan Qiu Hua
• Keywords: Classification, Content-Based Visual Retrieval, Visual Concept Detection

References

[1] DAVID G. LOWE, Distinctive Image Features from Scale-Invariant Keypoints, International Journal of Computer Vision 60(2), 91–110, (2004)

DOI: 10.1023/b:visi.0000029664.99615.94

[2] L. Duan*, D. Xu, I. W. Tsang and J. Luo, "Visual Event Recognition in Videos by Learning from Web Data," IEEE Trans. on Pattern Analysis and Machine Intelligence (T-PAMI), Accepted (In Press).

DOI: 10.1109/tpami.2011.265

[3] L. Duan*, I. W. Tsang, D. Xu and S. J. Maybank, "Domain Transfer SVM for Video Concept Detection," IEEE Int. Conf. on Computer Vision and Pattern Recognition (CVPR), 2009.

DOI: 10.1109/cvpr.2009.5206747

[4] K.-H. Liu, M.-F. Weng, C.-Y. Tseng, Y.-Y. Chuang, and M.-S. Chen. Association and temporal rule mining for post-filtering of semantic concept detection in video. IEEE Transactions on Multimedia, 10(2):240-251, February 2008.

DOI: 10.1109/tmm.2007.911826

[5] Mark J. Huiskes, Bart Thomee, Michael S. Lew: New trends and ideas in visual concept detection: the MIR flickr retrieval evaluation initiative. Multimedia Information Retrieval 2010: 527-536

DOI: 10.1145/1743384.1743475

[6] Nowak, S. and Dunker, P. (2009) Overview of the CLEF 2009 Large-scale Visual Concept Detection and Annotation Task, CLEF 2009 Workshop, Corfu, Greece

DOI: 10.1007/978-3-642-15751-6_10

[7] Arjan T. Setz and Cees G. M. Snoek, "Can Social Tagged Images Aid Concept-Based Video Search?," in Proceedings of the IEEE International Conference on Multimedia & Expo, New York, NY, USA, 2009, pp.1460-1463.

DOI: 10.1109/icme.2009.5202778

[8] K. E. A. van de Sande, T. Gevers, and C. G. M. Snoek. A comparison of color features for visual concept classification. In Proceedings of the ACM International Conference on Image and Video Retrieval, pages 141-149. Niagara Falls, Canada, July 2008.

DOI: 10.1145/1386352.1386376

[9] S. Lazebnik, C. Schmid, and J. Ponce. Beyond bags of features: Spatial pyramid matching for recognizing natural scene categories. In IEEE Conference on Computer Vision and Pattern Recognition, volume 2, pages 2169–2178, New York, USA, 2006.

DOI: 10.1109/cvpr.2006.68

[10] K. M. Borgwardt, A. Gretton, M. J. Rasch, H.-P. Kriegel, B. Sch¨olkopf and A. J. Smola. Integrating structured biological data by kernel maximum mean discrepancy. In ISMB, 2006.

DOI: 10.1093/bioinformatics/btl242

[11] Huiskes, M.J. and Lew, M. S. 2008. The MIR Flickr retrieval evaluation. In Proceeding of the 1st ACM international Conference on Multimedia information Retrieval (Vancouver, British Columbia, Canada, October 30 - 31, 2008). MIR '08. ACM, New York, NY, 39-43.

DOI: 10.1145/1460096.1460104