The effect is not as good as the latest BiRefNet model
#17
by jj11 - opened
The effect is not as good as the latest BiRefNet model
Our model excels on background removal and shows higher success rate over BirefNet model on our benchmark.
origubany changed discussion status to closed
Thank you for your feedback! While no model is perfect across all scenarios, we continually strive to improve performance. To support transparency and evaluation, we’ve created an open and diverse benchmark that you can explore here: https://github.com/Efrat-Taig/RMBG-2.0. We’d love to hear your thoughts or suggestions based on your testing.