BPT Enhancement Based on Syntactic and Semantic Criteria
This paper presents two enhancements for the creation and analysis of Binary Partition Trees (BPTs). Firstly, the classic creation of BPT based on colour is expanded to include syntactic criteria derived from human perception. Secondly, a method to include semantic information in the BPT analysis is shown thanks to the definition of the BPT Semantic Neighborhood and the introduction of Semantic Trees. Both techniques aim at bridging the semantic gap between signal and semantics following a bottom-up and a top-down approach, respectively.
KeywordsDissimilarity Measure Initial Partition Classic Creation Object Extraction Semantic Tree
Unable to display preview. Download preview PDF.
- 2.Ferran, C., Casas, J.R.: Object representation using colour, shape and structure criteria in a binary partition tree. In: International Conference on Image Processing (ICIP 2005), Genoa, vol. III, pp. 1144–1151. IEEE, Los Alamitos (2005)Google Scholar
- 3.Koffka, K.: Principles of Gestalt Psychology (1935)Google Scholar
- 4.Desolneux, A., Moisan, L., Morel, J.M.: Computational gestalts and perception thresholds. Journal of Physiology 97(2-3), 311–322 (2003)Google Scholar
- 6.Giró, X., Marqués, F.: Detection of semantic objects using description graphs. In: IEEE International Conference on Image Processing, ICIP 2005 (2005)Google Scholar
- 7.Moellic, P.A., Hede, P., Grefenstette, G., Mullet, C.: Evaluating content based image retrieval techniques with the one million images clic testbed. In: Proceedings of the International Conference on Pattern recognition and Computer Vision (2005)Google Scholar