ABSTRACT
It has been established that active learning is effective for learning complex, subjective query concepts for image retrieval. However, active learning has been applied in a concept independent way, (i.e., the kernel-parameters and the sampling strategy are identically chosen) for learning query concepts of differing <i>complexity</i>. In this work, we first characterize a concept's complexity using three measures: <i>hit-rate</i>, <i>isolation</i> and <i>diversity</i>. We then propose a multimodal learning approach that uses images' semantic labels to guide a <i>concept-dependent</i>, <i>active-learning</i> process. Based on the complexity of a concept, we make intelligent adjustments to the sampling strategy and the sampling pool from which images are to be selected and labeled, to improve concept learnability. Our empirical study on a $300$K-image dataset shows that concept-dependent learning is highly effective for image-retrieval accuracy.
- R. Agrawal, T. Imielinski, and A. Swami. Mining association rules between sets of items in large databases. Proceedings of ACM SIGMOD, 1993. Google ScholarDigital Library
- K. Barnard, P. Duygulu, and D. Forsyth. Exploiting text and image feature co-occurrence statistics in large datasets. Trends and Advances in Content-Based Image and Video Retrieval (To Appear), 2004.Google Scholar
- K. Barnard and D. Forsyth. Learning the semantics of words and pictures. In International Conference on Computer Vision, volume~2, pages 408--415, 2000.Google Scholar
- A. B. Benitez and S.-F. Chang. Image classification using multimedia knowledge networks. Proc. of the Int. Conf. on Image Processing, September 2003.Google ScholarCross Ref
- A. B. Benitez, J. R. Smith, and S.-F. Chang. Medianet: A multimedia information network for knowledge representation. Proceeding of the SPIE Conference on Internet Multimedia Management Systems, November 2000.Google ScholarCross Ref
- K. Brinker. Incorporating diversity in active learning with support vector machines. Proceedings of the Twentieth International Conference on Machine Learning, pages 59--66, August 2003.Google Scholar
- E. Chang and B. Li. Mega --- the maximizing expected generalization algorithm for learning complex query concepts. ACM Transaction on Information Systems, December 2003. Google ScholarDigital Library
- E. Y. Chang, K. Goh, and W.-C. Lai. On scalability of active learning for formulating query concepts. Workshop on Computer Vision Meets Databases (CVDB) in conjunction with ACM SIGMOD, June 2004. Google ScholarDigital Library
- R. Duda, P. Hart, and D. G. Stork. Pattern Classification. Wiley, New York, 2 edition, 2001. Google ScholarDigital Library
- A. K. Jain, R. P. Duin, and J. Mao. Statistical pattern recognition: A review. IEEE Trans. on Pattern Analysis and Machine Intelligence, 22(1):4--37, January 2000. Google ScholarDigital Library
- J. Jeon, V. Lavrenko, and R. Manmatha. Automatic image annotation and retrieval using cross-media relevance models. Proceedings of the 26th Annual International ACM-SIGIR Conference on Research and Development in Information Retrieval, August 2003. Google ScholarDigital Library
- C. Li, E. Chang, H. Garcia-Molina, and G. Wilderhold. Clindex: Approximate similarity queries in high-dimensional spaces. IEEE Transactions on Knowledge and Data Engineering (TKDE), 14(4), July 2002. Google ScholarDigital Library
- J. Li and J. Z. Wang. Automatic linguistic indexing of pictures by a statistical modeling approach. IEEE Transactions on Pattern Analysis and Machine Intelligence, 25(20), 2003. Google ScholarDigital Library
- Y. Lu, C. Hu, X. Zhu, H. Zhang, and Q. Yang. A unified framework for semantics and feature based relevance feedback in image retrieval systems. In Proc. of ACM Int. Conf. on Multimedia, pages 31--37, 2000. Google ScholarDigital Library
- Y. Mori, H. Takahashi, and R. Oka. Automatic words assignment to images based on image division and vector quantization. In Proc. of RIAO 2000: Content-Based Multimedia Information Access, Apr. 2000.Google Scholar
- S. Tong and E. Chang. Support vector machine active learning for image retrieval. Proc. of ACM Int. Conf. on Multimedia, pages 107--118, October 2001. Google ScholarDigital Library
- T. Westerveld. Image retrieval: Content versus context. Content-Based Multimedia Information Access, RIAO, pages 276--284, 2000.Google Scholar
- H. Zhang, Z. Chen, M. Li, and Z. Su. Relevance feedback and learning in content-based image search. WWW: Internet and Web Information Systems, 6(2):131--155, 2003. Google ScholarDigital Library
- X. S. Zhou and T. S. Huang. Unifying keywords and visual contents in image retrieval. IEEE Multimedia, 9(2):23--33, 2002. Google ScholarDigital Library
- L. Zhu, A. Rao, and A. Zhang. Theory of keyblock-based image retreival. ACM Trans. on Information Systems, 20(2):224--257, 2002. Google ScholarDigital Library
Index Terms
- Multimodal concept-dependent active learning for image retrieval
Recommendations
Support vector machine active learning for image retrieval
MULTIMEDIA '01: Proceedings of the ninth ACM international conference on MultimediaRelevance feedback is often a critical component when designing image databases. With these databases it is difficult to specify queries directly and explicitly. Relevance feedback interactively determinines a user's desired output or query concept by ...
Optimization on active learning strategy for object category retrieval
ICIP'09: Proceedings of the 16th IEEE international conference on Image processingActive learning is a machine learning technique which has attracted a lot of research interest in the content-based image retrieval (CBIR) in recent years. To be effective, an active learning system must be fast and efficient using as few (relevance) ...
Batch Mode Active Learning for Interactive Image Retrieval
ISM '14: Proceedings of the 2014 IEEE International Symposium on MultimediaIn content-based image retrieval, relevance feedback is an effective approach to narrow the semantic gap between low-level feature and high-level semantic interpretation by using user's feedback to judge the relevance images in the retrieval process. ...
Comments