´ Etude comparative de metriques pour l'´evaluation de la localisation d'objets par des boˆõtes englobantes.

This paper deals with a comparative study of metrics allowing the evaluation of results provided by object lo- calization algorithms. We particulary focus on localization by the bounding box representation. 26 metrics are studied in this paper. A protocol is presented for the creation of ground truths and synthetic results of localization algorithms. These synthetic results permit to simulate several errors (translation, scale errors. . . ) and to study the metrics behaviors face to the considered alterations. Experimental results illustrate the reliability of the different metrics face to different kinds of error.

[1]  William A. Yasnoff,et al.  Error measures for scene segmentation , 1977, Pattern Recognit..

[2]  Olivier D. Faugeras,et al.  Visual Discrimination of Stochastic Texture Fields , 1978, IEEE Transactions on Systems, Man, and Cybernetics.

[3]  E. D'Angelo,et al.  ROBIN Challenge Evaluation principles and metrics , 2006 .

[4]  Ph. Bolon,et al.  Evaluation quantitative d'images filtrées , 1997 .

[5]  M. Beauchemin,et al.  On the Hausdorff Distance Used for the Evaluation of Segmentation Results , 1998 .

[6]  L. Vinet.  Segmentation et mise en correspondance de régions de paires d'images stéréoscopiques , 1991 .

[7]  Hugues Benoit-Cattin,et al.  Scalable discrepancy measures for segmentation evaluation , 2002, Proceedings. International Conference on Image Processing.

[8]  Hélène Laurent,et al.  Unsupervised and Supervised Image Segmentation Evaluation , 2006 .

[9]  Qian Huang,et al.  Quantitative methods of evaluating image segmentation , 1995, Proceedings., International Conference on Image Processing.

[10]  Adel Hafiane,et al.  Caracterisation de Textures et Segmentation pour la recherche d'images par le contenu. (Texture characterization and Segmentation for Content Based Image Retrieval) , 2005 .

[11]  M. Basseville.  Distance measures for signal processing and pattern recognition , 1989 .

[12]  Jitendra Malik,et al.  A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics , 2001, Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001.

[13]  E. Baltsavias,et al.  3-D Building Reconstruction with ARUBA: A Qualitative and Quantitative Evaluation , 1997 .

[14]  Cordelia Schmid,et al.  The 2005 PASCAL Visual Object Classes Challenge , 2005, MLCW.