qinnzou

Results 24 comments of qinnzou

Thanks for your question. In the val_example.txt file, each raw has two images, where the first one is the image for test, and the second one is the ground truth....

The .png and the .bmp contain the same content, i.e., the crack ground truth. The jpg is generally a loss compression format for images that would bring uncerntainty for crack...

The precision-recall results can be calculated by using the edge detection benchmark of Berkeley. The codes can be found at https://www2.eecs.berkeley.edu/Research/Projects/CS/vision/bsds/

Yes. @zhuhaonan We will release the test results. @chenghd98 I did not know what GPU you used. The running speed reported in our paper is based on two TiTAN.

@lian666-ch The precision-recall results can be calculated by using the edge detection benchmark of Berkeley. The codes can be found at https://www2.eecs.berkeley.edu/Research/Projects/CS/vision/bsds/

large GPU memory is required.

@prinzROSE training from scratch