Semantic Intensity: Objects Contributions Towards Image Annotation
The fast-growing innovations within the field of digital media provide a platform where the sizes of the digital contents are expanding aggressively without proper management. Management of such kind of digital data needs proper techniques to facilitate easy retrieval of specific objects intelligently out of stored materials. Object detection and labelling with a proper term are one of the main issues within the field of multimedia. This has led to a problem of how these contents can be effectively managed. The management of these digital contents has not gotten as much attention as compared to the production and technology development. A lot of useful information goes to wastage due to the poor management of digital contents. This paper focusses on the Semantic Intensity (SI), which is an approach to arrange the identical object’s images in the dataset. SI can easily be calculated for each of the single object in the images bases on their polygon point representation. The SI-values represent the degree of contribution of each of the object towards semantics of the image. From the experimental work, our algorithm performance was good in terms of calculating the SI-value of the objects inside the image.
KeywordsSemantic intensity Annotated images LabelMe
- 4.Zhang, X., Jin, X., Li, L., Shen, D.: Discovering multilingual concepts from unaligned web documents by exploring associated images. In: Proceedings of the 22nd International Conference on WWW Companion and Steering Committee Republic and Canton of Geneva, Switzerland, pp. 173–174 (2013)Google Scholar
- 5.Aslam, N.: Semantic multimedia modeling and interpretation for search and retrieval. Ph.D. thesis, Middlesex University (2011)Google Scholar
- 7.Dayong, W., Steven, C.H.H., Ying, H.: Mining weakly labeled web facial images for search-based face annotation. In: International Conference on SIGIR 2011, pp. 1–10. ACM (2011)Google Scholar
- 8.Zhou, T.H., Wang, L., Shon, H.S., Lee, Y.K., Ryu, K.H.: Correlated multi-label refinement for semantic noise removal. In: Huang, D.-S., Zhang, X., Reyes García, C.A., Zhang, L. (eds.) ICIC 2010. LNCS, vol. 6216, pp. 309–316. Springer, Heidelberg (2010)Google Scholar
- 11.Li, X., Hu, W., Shen, C., Zhang, Z., Dick, A., Hengel, A.V.D.: A survey of appearance models in visual object tracking. J. ACM Trans. Intell. Syst. Technol. (TIST) 4(4), 1–48 (2013). Article No. 58Google Scholar