Giters

justusschock / shapenet

PyTorch implementation of "Super-Realtime Facial Landmark Detection and Shape Fitting by Deep Regression of Shape Model Parameters" predicting facial landmarks with up to 400 FPS

Home Page:https://shapenet.rtfd.io

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Training Question

iPsych opened this issue · comments

commented

Hello,
I found your answer in #31.
Pre-trained weight of Thermal faces are still not available?
I downloaded ThermalFace Database, and about to train.
Shall I duplicate config file of helen or some adjustment is required?

commented

Hi,

No the weights are still not available and I probably won't put them online in the future as well.

I think the values of the Helen config file should be fine for the training (maybe you don't need that much of rotation).

commented

@justusschock Hi,
Thank you for the quick reply.
I am trying to train by myself using thermalface DB. however the json structure is quite different from ibug.
Did you convert it using some code or tool?

commented

no, we didn't. The thermal DB also follows the ljson standard from Menpo and was loaded using this code: https://github.com/justusschock/shapedata/blob/master/shapedata/io.py#L54 (which is also taken from Menpo/ibug)