Hello. I just saw your data and demo from Facebook.
I think you should improve your mask more precisely if you want to get better results.
It is still ok but if you want to get really good result, you need to consider it.
Thank you for sharing your project.

Hi Nikhil,
thank you for the project. I reproduced the results, everything went well till the end. But I have got the original as a grayscale image at the end in the results folders. I don't know why. Your help will be appreciated in this regard.
Thanks,
Asif Ahmad

Hello! Hope your doing great. I just wanted to reproduce the results. Every thing was going smoothly then this error prompted at the end in predict.py file. I am trying this on Google Colab Details of the error are given below, kindly check.

The code line to which the error points is:

The folders, files and results are inside another folder named as Filess and I have changed the directory to the that too.

Any help would be greatly appreciated. Thanks!

I have followed your article to train a model based on my custom data set successfully. Now what I want is to use it for the segmentation of 1 picture per iteration. Any kind of help in this regard would be appreciated.

Hi there,

When I run the predict.py script I get this error" TypeError: '<' not supported between instances of 'function' and 'str'". I was wondering if you have any thoughts on this issue? Thanks