Face Conversion Integrity
 

We've been converting faces from 3DMeNow into PeopleMaker and then Vizard for a pretest of facial attractiveness. We've noticed that the final faces in Vizard look somewhat different from how they look in 3DMeNow (lower resolution, stronger shadowing, other minor differences).

Are there ways to minimize the differences between the 3DMeNow and PeopleMaker faces?

Nick

Hi Nick,

The resolution you export in 3DMeNow should be the same when you import into PeopleMaker. When you export in 3DMeNow, make sure you select the appropriate model detail.

Setting the appearance of the face to DECAL should remove the shadowing. Example:What are the other differences you notice?

When I export from 3DMeNow, whatever detail level I set it at, it seems to be exporting a low detail version in VRML format. Is the export setting set elsewhere?

Is there any documentation on PeopleMaker yet?

Nick

There are two details level settings available in 3DMeNow. One refers to the level of photo-matching detail, and the other refers to the level of polygonal complexity. They are entirely independent: a model built with high photo-matching detail can be exported as low polygons and vice-versa. The software refers to the polygonal detail level as the "subdivision" level ranging between 0, 1, and 2 (0 being the fewest polygons). Let me know if that doesn't answer your first question.

Regarding PeopleMaker documentation, no, it is still an alpha release utility that ships with Vizard in a relatively unsupported fashion. I can't promise when that is due to change.