Skip to main content
Account

Table 1 User-defined parameters of the energy function \(\mathbf E \), together with their description, values interval and default value

From: Multi-view Performance Capture of Surface Details

Parameter

Description

Values int.

Def.

\(w_{reg}\)

Regularization weight

[0, 1]

\(5e^{-7}\)

\(w_{temp}\)

Temporal weight

[0, 1]

\(1e^{-7}\)

\(\sigma _s\)

Standard deviation, \(\forall G_s\)

\((0,\infty )\), [mm]

5

\(T_{qt}\)

Quad-tree depth threshold

\([0,log_2(I^{W,H}) ]\)

9

\(T_{fuse}\)

Color similarity for fusion threshold

\([0,\infty )\)

0.05

\(T_{color}\)

Color similarity threshold

\([0,\infty )\)

0.15

\(T_{dist}\)

Threshold on pixel distance

\([0,\infty )\), [px]

30

\({\varDelta }_d\)

Max geodesic distance

\([0,\infty )\), \([\# edges]\)

2

  1. The quad-tree depth threshold \(T_{qt}\) and color similarity threshold \(T_{fuse}\) are used to subdivide the input 2D images as described in Sect. 4.2 and find the Image Gaussians \(G_i\). \(T_{qt}\) can be at most the logarithm in base 2 of the minimum among the image width \(I^W\) and the image height \(I^H\), expressed as \(log_2(I^{W,H})\). The thresholds on color \(T_{color}\) and pixel distance \(T_{dist}\) between Surface \(G_s\) and Image Gaussians \(G_i\) define for each \(G_s\) the subset of \(G_i\) against which the similarity function is evaluated, excluding less contributing Image Gaussians (see Sect. 5.4)