23

Figure 3.3 : Interactive painting-based segmentation of human head: the density
volume 3.3(a), restricting painting to thresholded bones 3.3(b), user painting with
spherical brush 3.3(c), and labeling result after graph-cut segmentation 3.3(d).
3.3 Tri-Iinear contours with Interactive Segmentation
We developed a simple painting-based segmentation interface to test and demonstrate
our proposed representation. Our segmentation interface considers the input as a raw
density map. The first step involves choosing one or more thresholds and generating
material masks as described in Section 3.2. This step corresponds to an initial stage of
segmentation by thresholding. Once these threshold materials have been generated,
the user has the choice to turn the materials into visible or non-visible regions for
painting. For example, when working on a human appendage, we can first turn the
flesh invisible to paint and segment out the individual bones from the threshold level
that corresponds to the bones. Figure 3.3 is an example of this process.
For painting, we provide the user with a spherical brush whose radius is vari-
able. To position the sphere, we construct a ray that is perpendicular to the viewing
plane and passes through the mouse position. The brush is positioned at the closest
intersection point between the projecting ray the the visible regions.
The user can then hold down a button to indicate the start of a sweep. The
sweep ends when the user releases the button. The sweep is constructed as a union
of spheres, where the center of each sphere corresponds to the position of the mouse
at some sampled instance during the the sweep. We compute a scalar field Ssphere1
More intriguing information
1. The name is absent2. Feature type effects in semantic memory: An event related potentials study
3. The Effects of Attendance on Academic Performance: Panel Data Evidence for Introductory Microeconomics
4. Elicited bid functions in (a)symmetric first-price auctions
5. The name is absent
6. Demand Potential for Goat Meat in Southern States: Empirical Evidence from a Multi-State Goat Meat Consumer Survey
7. A methodological approach in order to support decision-makers when defining Mobility and Transportation Politics
8. Regional specialisation in a transition country - Hungary
9. The name is absent
10. Olfactory Neuroblastoma: Diagnostic Difficulty