Automatic Images Annotation Extension Using a Probabilistic Graphical Model

Abdessalem Bouzaieni (1) , Salvatore Tabbone (1) , Sabine Barrat (2)

1 QGAR - Querying Graphics through Analysis and Recognition
2 LIFAT - Laboratoire d'Informatique Fondamentale et Appliquée de Tours

With the fast development of digital cameras and social media image sharing, automatic image annotation has become a research area of great interest. It enables indexing, extracting and searching in large collections of images in an easier and faster way. In this paper, we propose a model for the annotation extension of images using a probabilistic graphical model. This model is based on a mixture of multinomial distributions and mixtures of Gaussians. The results of the proposed model are promising on three standard datasets: Corel-5k, ESP-Game and IAPRTC-12.

Mots clés: image annotation; annotation extension; probabilistic graphical models; Gaussian mixtures; visual characteristics; textual characteristics.

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]

Abdessalem Bouzaieni, Salvatore Tabbone, Sabine Barrat. Automatic Images Annotation Extension Using a Probabilistic Graphical Model. CAIP 2015 - 16th International Conference on Computer Analysis of Images and Patterns, Sep 2015, Valetta, Malta. pp.579-590, ⟨10.1007/978-3-319-23117-4_50⟩. ⟨hal-01254926⟩