Documentos de conferencia

URI permanente para esta colección

https://hdl.handle.net/10784/28291

Examinar

Examinando Documentos de conferencia por Materia "Algorithms"

Mostrando 1 - 2 de 2
  • No hay miniatura disponible
    Ítem
    Parametric curve reconstruction from point clouds using minimization techniques
    (2013-01-01) Ruiz, O.E.; Cortés, C.; Aristizábal, M.; Acosta, D.A.; Vanegas, C.A.; Universidad EAFIT. Departamento de Ingeniería de Procesos; Desarrollo y Diseño de Procesos
    Curve reconstruction from noisy point samples is central to surface reconstruction and therefore to reverse engineering, medical imaging, etc. Although Piecewise Linear (PL) curve reconstruction plays an important role, smooth (C1-, C2-,?) curves are needed for many applications. In reconstruction of parametric curves from noisy point samples there remain unsolved issues such as (1) high computational expenses, (2) presence of artifacts and outlier curls, (3) erratic behavior of self-intersecting curves, and (4) erratic excursions at sharp corners. Some of these issues are related to non-Nyquist (i.e. sparse) samples. In response to these shortcomings, this article reports the minimization-based fitting of parametric curves for noisy point clouds. Our approach features: (a) Principal Component Analysis (PCA) pre-processing to obtain a topologically correct approximation of the sampled curve. (b) Numerical, instead of algebraic, calculation of roots in point-to-curve distances. (c) Penalties for curve excursions by using point cloud to - curve and curve to point cloud. (d) Objective functions which are economic to minimize. The implemented algorithms successfully deal with self - intersecting and / or non-Nyquist samples. Ongoing research includes self-tuning of the algorithms and decimation of the point cloud and the control polygon.
  • No hay miniatura disponible
    Ítem
    Statistical tuning of adaptive-weight depth map algorithm
    (SPRINGER, 2011-01-01) Hoyos, Alejandro; Congote, John; Barandiaran, Inigo; Acosta, Diego; Ruiz, Oscar; Universidad EAFIT. Departamento de Ingeniería de Procesos; Desarrollo y Diseño de Procesos
    In depth map generation, the settings of the algorithm parameters to yield an accurate disparity estimation are usually chosen empirically or based on unplanned experiments. A systematic statistical approach including classical and exploratory data analyses on over 14000 images to measure the relative influence of the parameters allows their tuning based on the number of bad-pixels. Our approach is systematic in the sense that the heuristics used for parameter tuning are supported by formal statistical methods. The implemented methodology improves the performance of dense depth map algorithms. As a result of the statistical based tuning, the algorithm improves from 16.78% to 14.48% bad-pixels rising 7 spots as per the Middlebury Stereo Evaluation Ranking Table. The performance is measured based on the distance of the algorithm results vs. the Ground Truth by Middlebury. Future work aims to achieve the tuning by using significantly smaller data sets on fractional factorial and surface-response designs of experiments. © 2011 Springer-Verlag.