R CM wrote: > you must make the sum of the areas of each triangle. > and for the mathematical volume, > i think it's more difficult... > perhaps with a octree method and test for each cell if > it's indside or outside the shape... > ( a non-trivial test ;-) ) > > anyone have a better solution ?
For convex hull, take any point inside a volume, and compute sum of volumes of 4-point pyramids created by each triangle of hull and this point inside. For open areas or models in which triangles cross each other, it is absolutely non-trival (in addition for needing a definition what volume means in such case). For nonconvex hulls, you have to dissect them somehow. It is generally compute-intensive. For some references you can try out http://users.comlab.ox.ac.uk/joe.pitt-francis/convex_decomp_talk.ps.gz but try out too look at google for 'tetrahedralization of three dimensional polyhedra' Artur =========================================================================== To unsubscribe, send email to [EMAIL PROTECTED] and include in the body of the message "signoff JAVA3D-INTEREST". For general help, send email to [EMAIL PROTECTED] and include in the body of the message "help".
