Knowledge-driven understanding of images in comic books

Analyse d'images de bande dessinée guidée par une connaissance à priori

Christophe Rigaud (1, 2) , Clément Guérin (1) , Dimosthenis Karatzas (2) , Jean-Christophe Burie (1) , Jean-Marc Ogier (1)

1 L3I - Laboratoire Informatique, Image et Interaction - EA 2118
2 CVC - Computer Vision Center (Centre de visio per computador)

Document analysis is an active field of research, which can attain a complete understanding of the semantics of a given document. One example of the document understanding process is enabling a computer to identify the key elements of a comic book story and arrange them according to a predefined domain knowledge. In this study, we propose a knowledge-driven system that can interact with bottom-up and top-down information to progressively understand the content of a document. We model the comic book’s and the image processing domains knowledge for information consistency analysis. In addition, different image processing methods are improved or developed to extract panels, balloons, tails, texts, comic characters and their semantic relations in an unsupervised way.

Mots clés: Document understanding; Comics analysis; Expert system; Knowledge representation

Domaines: Traitement des images [eess.IV]; Recherche d'information [cs.IR]; Informatique [cs]

Christophe Rigaud, Clément Guérin, Dimosthenis Karatzas, Jean-Christophe Burie, Jean-Marc Ogier. Knowledge-driven understanding of images in comic books. International Journal on Document Analysis and Recognition (IJDAR), 2015, 18, pp.1433-2833. ⟨10.1007/s10032-015-0243-1⟩. ⟨hal-01144856⟩