Proceedings chapter

Hierarchical long-term learning for automatic image annotation

Presented at Genova (Italy), Dec 5-7
  • Lecture Notes in Computer Science; 4816
Publication date2007

This paper introduces a hierarchical process for propagating image annotations throughout a partially labelled database. Long-term learning, where users' query and browsing patterns are retained over multiple sessions, is used to guide the propagation of keywords onto image regions based on low-level feature distances. We demonstrate how singular value decomposition (SVD), normally used with latent semantic analysis (LSA), can be used to reconstruct a noisy image-session matrix and associate images with query concepts. These associations facilitate hierarchical filtering where image regions are matched based on shared parent concepts. A simple distance-based ranking algorithm is then used to determine keywords associated with regions.

Citation (ISO format)
MORRISON, Donn Alexander, MARCHAND-MAILLET, Stéphane, BRUNO, Eric. Hierarchical long-term learning for automatic image annotation. In: 2nd International Conference on Semantic and Digital Media Technologies, SAMT 2007 : proceedings. Genova (Italy). [s.l.] : Springer, 2007. p. 28–40. (Lecture Notes in Computer Science) doi: 10.1007/978-3-540-77051-0_3

Technical informations

Creation03/06/2015 5:12:14 PM
First validation03/06/2015 5:12:14 PM
Update time03/14/2023 10:59:11 PM
Status update03/14/2023 10:59:11 PM
Last indexation08/29/2023 3:11:18 PM
All rights reserved by Archive ouverte UNIGE and the University of GenevaunigeBlack