Hi there, I wanted to share some animation I did recently using human models created with PhotoScan.
We shot several actors holding frozen poses and built them up as models using PhotoScan. These were then placed into 3D scenes where we could light, animate (the camera) and render them.
I spent about a week coming up with a good workflow for improving the quality of the models and getting the best texture results. In the end I settled on using zBrush to retopologize and clean the mesh. Then I could create new UVs and bring the model back into PhotoScan to project textures. I also decided to output one image at a time from PhotoScan and then combine them into a single texture in Photoshop. The texture merging inside PhotoScan was resulting in quite a lot of blurriness.
Here's some of the shots we produced:
https://vimeo.com/60765544