My previous workflow for 3D reference positioning just kinda… quit working. I don’t know why. I think it was around the time I upgraded to CS6, but downgrading to CS5 didn’t fix it. I flailed around for a while, thinking crazy thoughts like “maybe Blender can do it”, tried a photo-match feature in Sketchup, and then after giving up for a while I thought about how Sketchup’s photo-match feature works and came up with a simple way that works better than my previous workflow.
I think I may have used this for the last panel of 3D ref in Rita, but hey, it’ll be there when I need it for the next time.
- Expand subdivision into a mesh, export an .obj from Silo.
- Load rough page into Photoshop. Convert from indexed to RGB.
- 3d->new layer from 3d file.
- Keeping the object’s ‘current view’ selected in the 3D window, switch to the move tool.
- Rotate and transpose the CAMERA VIEW to match the sketch. DO NOT MOVE THE MODEL AROUND. Keep the model in the center of the 3d space.
- Also futz with the field of view in the 3D panel to match whatever exaggerated perspective I’ve done in my sketch.
- Export a transparent png, drop into AI, draw.