english Icono del idioma   español Icono del idioma  

Por favor, use este identificador para citar o enlazar este ítem: https://hdl.handle.net/20.500.12008/30466 Cómo citar
Título: The whole and the parts : The MDL principle and the a-contrario framework
Autor: Grompone von Gioi, Rafael
Ramírez Paulino, Ignacio
Randall, Gregory
Tipo: Preprint
Palabras clave: Model selection, Structure detection, MDL, A-contrario framework, Non accidentalness principle, NFA, Polygonal approximation, Line segment detection
Fecha de publicación: 2021
Resumen: This work explores the connections between the Minimum Description Length (MDL) principle as developed by Rissanen, and the a-contrario framework for structure detection proposed by Desolneux, Moisan and Morel. The MDL principle focuses on the best interpretation for the whole data while the a-contrario approach concentrates on detecting parts of the data with anomalous statistics. Although framed in different theoretical formalisms, we show that both methodologies share many common concepts and tools in their machinery and yield very similar formulations in a number of interesting scenarios ranging from simple toy examples to practical applications such as polygonal approximation of curves and line segment detection in images. We also formulate the conditions under which both approaches are formally equivalent.
Editorial: arXiv
EN: Computer Science (cs.CV-Computer Vision and Pattern Recognition), arXiv:2112.06853, Dec. 2021, pp. 1-32.
Citación: Grompone von Gioi, R., Ramírez Paulino, I. y Randall, G. The whole and the parts : The MDL principle and the a-contrario framework [Preprint]. Publicado en : Computer Science (cs.CV-Computer Vision and Pattern Recognition), 2021, pp. 1-32. arXiv:2112.06853.
Licencia: Licencia Creative Commons Atribución - No Comercial - Compartir Igual (CC - By-NC-SA 4.0)
Aparece en las colecciones: Publicaciones académicas y científicas - Instituto de Ingeniería Eléctrica

Ficheros en este ítem:
Fichero Descripción Tamaño Formato   
GRR21.pdfPreprint5,75 MBAdobe PDFVisualizar/Abrir


Este ítem está sujeto a una licencia Creative Commons Licencia Creative Commons Creative Commons