ONLINE KERNEL LEARNING FOR INTERACTIVE RETRIEVAL IN DYNAMIC IMAGE DATABASES

Philippe-Henri Gosselin (1)

1 MIDI - Multimedia Indexation and Data Integration

In this paper, we propose a system for interactive image retrieval in dynamic databases, where images are regularly added or removed. In order to handle this, we propose a method that tunes itself accord- ing to user labels. The framework we propose is based on visual dictionaries, with the specificity that the dictionaries are built online, during retrieval sessions. In other words, each user has its own vi- sual dictionary, as opposed to usual approaches where all users share the same visual dictionary. In order to create theses dictionaries, we propose a method based on kernel functions. This method iteratively selects base kernels from a large base kernel pool, where each base kernel is related to a low-level descriptor such as color or texture. This learning process is performed in real time, and the classifica- tion of the database is faster than usual techniques since only rele- vant features for the current query are used. Experiments are carried out on a generalist database, and show the ability of the method to build effective kernels with few labels.

Domains: Machine Learning [stat.ML]; Multimedia [cs.MM]

Philippe-Henri Gosselin. ONLINE KERNEL LEARNING FOR INTERACTIVE RETRIEVAL IN DYNAMIC IMAGE DATABASES. IEEE International Conference on Image Processing, Sep 2012, Orlando, United States. ⟨hal-00753154⟩