search

patrikhuber / eos

A lightweight 3D Morphable Face Model library in modern C++
Apache License 2.0
1.89k stars 596 forks source link

about model boundary vertices #345

Closed rlczddl closed 1 year ago

rlczddl commented 1 year ago

I wonder if the calculated model boundary vertices are not uniquely, is this normal?Here is my result on BFM07 model: image

rlczddl commented 1 year ago

I tested the fit-model example and found the same result. That is to say, there are multiple 3d contour points corresponding to the same 2d contour point, is that so? image

patrikhuber commented 1 year ago

Hi! Sorry for the late reply. To be honest, I'm not 100% sure - the code should give you the authoritative answer. But from my memory, I think that you might be right, and multiple 3D contour points can map to the same 2D contour point. I don't think that's an issue though - it may actually be useful for fairly low resolution model meshes, where there isn't a great 2D-3D correspondence for a particular 2D contour point.

rlczddl commented 1 year ago

Thanks for your reply,I see.