Skip to Main content Skip to Navigation
Conference papers

Extraction et comparaison d'information saillante : Pose favorable et image 2D révélatrice d'un objet 3D

Marie Pelissier Combescure 1 Géraldine Morin 1 Sylvie Chambon 1
1 IRIT-REVA - Real Expression Artificial Life
IRIT - Institut de recherche en informatique de Toulouse
Abstract : The objective of this paper is to quantify the information relative to a 3D object contained in a 2D image, by estimating, on the one hand, the interest of its pose and, on the other hand, the interest of the 2D observed representation. The goal is to extract, from these two different and complementary modalities, the essential information belonging to the object in the image. This information is computed by matching salient elements between the 2D image and the3D mesh. For this detection step, we use a detector between2D and 3D, based on the curvilinear saliency, that ensures the best possible repeatability for salient elements in different modalities. We are thus able to identify the pose that best highlights the 3D object based on the detected points of interest. Using a similar approach, we also quantify if the 2D image that contains this object highlights it. In particular, given a set of poses and images, we are able to order these poses and images according to their saliency : the interest of the pose and the relevance of the 2D image.
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-03339730
Contributor : Ccsd Sciencesconf.Org Connect in order to contact the contributor
Submitted on : Thursday, September 9, 2021 - 4:08:15 PM
Last modification on : Tuesday, October 19, 2021 - 2:24:25 PM

File

Orasis_version_finale.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03339730, version 1

Citation

Marie Pelissier Combescure, Géraldine Morin, Sylvie Chambon. Extraction et comparaison d'information saillante : Pose favorable et image 2D révélatrice d'un objet 3D. 18èmes journées francophones des jeunes chercheurs en vision par ordinateur (ORASIS 2021), Centre National de la Recherche Scientifique [CNRS]; Equipe REVA, IRIT : Institut de Recherche en Informatique de Toulouse., Sep 2021, Saint Ferréol, France. pp.1-9. ⟨hal-03339730⟩

Share

Metrics

Record views

47

Files downloads

18