songheng

1. You can train 192 or 168 points - model with this set of tools, for more info, read the code. 2. You can train on other dataset, generally, after...

Current network could work with image of other size. Image of large size will be more accurate, but the training speed will consume more time, images of small size will...

Suggest you check and try to modify the code by yourself, or you may wait, for I am planning to release a new version to solve these problems recently.

We plan to release new version next month. App will train model on Ubuntu 18 and 16, not windows.

check and change parameters in script directly.

请查一下运行的环境，tensorflow的版本，一定要用tensorflow 1.13 及以上，版本低，报错。 可以参考 README-ZH.md的中文说明，“”依赖“”部分。

现在的模型是针对脸部的，针对眼部，可采用级联法，结合传统算法。dlib库实现了face landmarks传统算法训练做眼部特征点精检。

Mobilenet (model) comsumes least computer resources among these provided in this tool set, and this model is able to run on embedded platform smoothly.

Tools should run on Ubuntu 16 or up, do not work on windows, do you run on windows?