The problem with 3D scanning, even at the high end, is that you wind up with incredibly dense geometry that doesn't lend itself to animation very well. I know here where I work, we always retopologize 3D scans to create an efficient mesh that has good polygon flow, crucial for getting nice deformations in animation.

A better option, if cost and simplicity is a concern, is to shoot photos and then simply map the images as textures on an existing mesh. You can use Poser or the free MakeHuman programs to shape a mesh close to the desired figure and features.

If you can model or sculpt and want to spend more time, use a program like 3D-Coat to model a mesh based on photographs. You can project the photos directly on the mesh in 3DC while painting it...an awesome feature. 3DC even has a nice sculpting feature called Voxel Sculpting, which is a lot like working with clay, and you can easily retopo the sculpt into a usable mesh (3DC has robust auto- and manual retopo tools that rea easy to use.) You may also use Mudbox or ZBrush to do the same; I prefer 3DC because IMO it's less expensive, has more features, and is easier to learn. But all three programs are very capable and used in the gaming and film industries.

Bear in mind that it's not enough to just have a model; you'll also need to rig and weight it, which you cannot do in any of the programs described above. But that's a whole other topic.

I use a program called 3dCoat...slick as snot.... when it works right. I have not had the need to use many 3d scans yet but may in the near future.