patrikhuber / 4dface

Real-time 3D face tracking and reconstruction from 2D video

Home Page:https://www.4dface.io

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

please help me, thank you

yyssmm opened this issue · comments

std::tie(mesh, rendering_params) = fitting::fit_shape_and_pose( ..........), you have fitted the pose, shape model, and expression blendshapes to landmarks in an iterative way using this function, and obtained a merged mesh, but why you merge the isomap and shape_coefficients again using the following two functions:
const Mat merged_isomap = isomap_averaging.add_and_merge(isomap);
shape_coefficients = pca_shape_merging.add_and_merge(shape_coefficients);

I think you are confusing between running on one image, and merging between different images.

Please adhere to better guidelines for posting issues in the future if you'd like to receive help - a clear title, description and reproducible case would be a good start. Thanks.