New descriptors for image and video indexing

Patrick Gros (1) , Ronan Fablet (2, 1) , Patrick Bouthemy (1)

1 IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires
2 SC - Département Signal et Communications

In this chapter, we present new descriptors that were specifically developed for content-based image and video indexing and retrieval. The first ones are an extension of the descriptors proposed by Florack et al. and Schmid to the case of color images. They are suited for object recognition in still images. The second ones describe the global level of motion activity in an image sequence, and are a good complement to the traditional color and texture descriptors.

Domaines: Traitement du signal et de l'image [eess.SP]

Patrick Gros, Ronan Fablet, Patrick Bouthemy. New descriptors for image and video indexing. State-of-the-art in content-based image and video retrieval, Kluwer Academic Publishers, pp.213 - 234, 2001, Computational imaging and vision, 1-4020-0109-6. ⟨hal-02283252⟩