DengPingFan / SINet

Camouflaged Object Detection, CVPR 2020 (Oral)

Home Page:http://dpfan.net/Camouflage

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Loss

jiangxinhao2020 opened this issue · comments

weit = 1+5*torch.abs(F.avg_pool2d(mask, kernel_size=31, stride=1, padding=15)-mask)
您好,您论文使用了F3Net的损失函数,请问kernel_size是否是超参数,31是如何选取的,能否选取51,15,9等数字。感谢解答!

这个参数是继承了原始F3Net模型设定:https://github.com/weijun88/F3Net/blob/eecace3adf1e8946b571a4f4397681252f9dc1b8/src/train.py#L19

你可以根据不同情况选取其他大小