Requesting open-source training code
#41
by demo001s - opened
I tested that removing background from photos often results in incomplete portraits or hard translucent edges. So could you please train your company's code so that I can use my own dataset to train it again.
The data we trained the model did not include portraits, it is likely that you need to retrain the model on data adapted and specific to your problem. For the training code, contact the Bria sales/solutions team and you can purchase the model and training code through our platform.