REALY

REALY copied to clipboard

will open the training code?

7

统一拓扑结构得到了约2000个不同性别、年龄、种族的人脸mesh，在此基础上，构建了一个全头人脸3DMM并命名为HIFI3D++. all these data preprocess script will be release?

foocker

Hey, where to download REALY_HIFI3D_keypoint and REALY_scan_region？

1

RujingD

Shape's jaw fit abnormally

3

This work is wonderful!! And I also paid attention to your previous related work https://github.com/tencent-ailab/hifi3dface, I directly replaced hifi3dface/3DMM/files/HIFI3D++.mat with hifi3dface/3DMM/files/HIFI3D++.mat The following error occurred when running: f_sigma0 = (paras)...

Allen-lz

Hello, is it somehow possible to convert a mesh in FLAME topology into a mesh in HIFI3D++ topology?

ImistroI

Hi, I recently wanted to register some 3dMD data, similar to Headspace data. I read your paper, but it is a bit difficult to reproduce. Could you please share with...

FortisCK

Face region mask

1

Thank you for release your awesome project. I noticed you released the HIFI3D++ 3DMM file. Because I am not a university employee, so I can't complete the dataset user agreement...

slime898

this project can be containerize and deployed on docker and can be started in isolated environment using docker

rituisboy

情况是这样，如果GT模型h和预测模型p分辨率差很多，比如GT模型一般可以用扫描仪获得，顶点数远远大于预测的模板模型。看论文bICP应该查找的也是一对一的点对，也就是最终计算的点对数是预测模板p的个数，这种评价方法很正确。我想请问，如果这个时候要用gICP，计算p->h方向比h->p方向应该更能反应真实的误差吧？我理解h->p方向GT模型h上会有很多点找到了预测模型p上面的同一点。

rlczddl

DECA evaluation with REALY

3

Hi. Thanks for sharing great works! I am YJHong and currently developing my own 3d face reconstruction based on DECA. After I've read evaluation protocol written in README, I have...

yjhong89