Open liuppboy opened 5 years ago
Hi, when I run the demo, I find that it does not add mean experission when computing vertices from the ground truth shape and expression parameters:
vertices = self.model['shapeMU'] + self.model['shapePC'].dot(shape_para) + self.model['expPC'].dot(exp_para)
While for 3DDFA and other projects, they all add mean experission, like this:
vertices = self.model['shapeMU'] + self.model['expMU'] + self.model['shapePC'].dot(shape_para) + self.model['expPC'].dot(exp_para)
It does not make much difference. I am just a little confused about this.
Hi, when I run the demo, I find that it does not add mean experission when computing vertices from the ground truth shape and expression parameters:
vertices = self.model['shapeMU'] + self.model['shapePC'].dot(shape_para) + self.model['expPC'].dot(exp_para)
While for 3DDFA and other projects, they all add mean experission, like this:
vertices = self.model['shapeMU'] + self.model['expMU'] + self.model['shapePC'].dot(shape_para) + self.model['expPC'].dot(exp_para)
It does not make much difference. I am just a little confused about this.