Training a new model
Hi,
Thank you for the code! It has been very useful to me. But I have run into a snag and I was hoping you can help.
I have been trying to train my own model with your code. I have a dataset of around 65000 images with 49 landmark points for each image.
While setting the parameters, I set "global_params.landmark_num" in InitializeGlobalParam() to 49. But the execution always crashes while loading the images at around the 16000 image mark or 30000 mark with the error - OpenCV Out of Memory exception.
However, if I leave the landmark_num to 68 the training completes successfully. But then if I try to Test the model or run the Demo mode, the program always crashes (which I think is what should happen since the global parameter claims there are 68 points but in reality there are only 49).
It would be really good if you could help me out a little.
Hi, thanks for your interest.
So you are using the C++ project, right? Please give me some time to check it. Because i suppose the C++ codes is not aimed to support different number of loanmarks.
Best,
Jianwei
On Tue, Sep 8, 2015 at 10:59 PM, sidc9 [email protected] wrote:
Hi,
Thank you for the code! It has been very useful to me. But I have run into a snag and I was hoping you can help.
I have been trying to train my own model with your code. I have a dataset of around 65000 images with 49 landmark points for each image.
While setting the parameters, I set "global_params.landmark_num" in InitializeGlobalParam() to 49. But the execution always crashes while loading the images at around the 16000 image mark or 30000 mark with the error - OpenCV Out of Memory exception.
However, if I leave the landmark_num to 68 the training completes successfully. But then if I try to Test the model or run the Demo mode, the program always crashes (which I think is what should happen since the global parameter claims there are 68 points but in reality there are only 49).
It would be really good if you could help me out a little.
— Reply to this email directly or view it on GitHub https://github.com/jwyang/face-alignment/issues/23.
Yes I'm using the C++ project. Thank you for the prompt response!
ok, I see, I will help to figure out why it did not work after changing the number of landmarks.
On Wed, Sep 9, 2015 at 10:58 PM, sidc9 [email protected] wrote:
Yes I'm using the C++ project. Thank you for the prompt response!
— Reply to this email directly or view it on GitHub https://github.com/jwyang/face-alignment/issues/23#issuecomment-139102214 .
Hi Jianwei,
I was wondering if you had any insights on the problem?
I am very sorry that I am a little busy these days, I have not yet found the causes. Maybe you can also send email to Yuanxun Wang, his email is [email protected]. He is the author of C++ codes.
@sidc9 I am very sorry that I am a little busy these days, I have not yet found the causes. Maybe you can also send email to Yuanxun Wang, his email is [email protected]. He is the author of C++ codes.