We have recently updated the models used for inference within FraD.
From this update onward, we are employing a model pre-trained on Imagenet and specifically trained for fracture surfaces (transfer learning).
We are committed to staying at the forefront of technological advancements and reflecting these innovations in our development efforts to enhance the performance of FraD.
Should you have any questions or concerns regarding the results produced by FraD, please do not hesitate to contact us. We appreciate your continued support.
