CVSL Logo
FrancaisHome
AboutPeopleResearchPublicationsEventsProfile
About
Publications

 

 

 

CERVIM

REPARTI

MIVIM

Detection of Multi-part Objects by Top-down Perceptual Grouping


Vénérée Randrianarisoa, Jean-François Bernier and Robert Bergevin


Abstract - In this paper, a top-down approach based on perceptual grouping is proposed for multi-part objects detection. The abstract conceptual category of multi-part objects is formalized by a set of global criteria. These criteria will enable the evaluation of the segmentation quality in order to determine if the whole grouping is perceptually significant and if it has a good perceptual shape. A new cognitive vision methodology, called SAFE (Subjectivity And Formalism Explicitly), is presented. Its goal is to help identify the proper global criteria and to validate the judgment derived from formal calculations of these criteria by human judgment.

download document

Bibtex:

@inproceedings{Randrianarisoa505,
    author    = { Vénérée Randrianarisoa and Jean-François Bernier and Robert Bergevin },
    title     = { Detection of Multi-part Objects by Top-down Perceptual Grouping },
    booktitle = { CRV 2005 },
    pages     = { 536-543 },
    address   = { Victoria University, British Columbia, Canada },
    year      = { 2005 },
    month     = { 9-11 May },
    journal   = { Second Canadian Conference on Computer and Robot Vision (CRV 2005) },
    keywords  = { Perceptual grouping, top-down processing, methodology, object detection },
    language  = { English }
}

Last modification: 2005/03/09 by veneree

     
   
   

©2002-. Computer Vision and Systems Laboratory. All rights reserved