Deep3DFaceReconstruction icon indicating copy to clipboard operation
Deep3DFaceReconstruction copied to clipboard

Why using ResNet-50 to regress the target vector instead of ResNet-101?

Open HhansssW opened this issue 4 years ago • 2 comments

Hi! Could you please give me some information about the reason choosing ResNet-50 ? Since I saw from the 3DMM-CNN that they used ResNet-101 to regress 198 3DMM parameters which is less than that in your work (257 parameters). Thank you so much.

HhansssW avatar Jun 09 '21 08:06 HhansssW

For the first question, we did not use ResNet-101 because it is kind of heavy and its performance is only slightly better compare to ResNet-50 in our experiment.

For the second question, we follow MoFA to set the dimension of 3DMM parameters in order to compare with it fairly.

YuDeng avatar Jun 11 '21 02:06 YuDeng

For the first question, we did not use ResNet-101 because it is kind of heavy and its performance is only slightly better compare to ResNet-50 in our experiment.

For the second question, we follow MoFA to set the dimension of 3DMM parameters in order to compare with it fairly.

Thanks ! ;)

HhansssW avatar Jun 14 '21 02:06 HhansssW