excllent123

@Shuaibishay Have you compared the result with A+? The result is bad compared with A+.and train takes a week in i7-8700 under div2k dateset.

直接加载模型就行，多GPU训练的，可能需要对模型进行修改后加载

https://blog.csdn.net/hardbird123/article/details/80549815 参考这个修改就行

你好，对于单目结构光，已经获得绝对相位、以及相应的内参、外参，如何获取depth? 可否分享一个代码？

Hi Could you share test file and weight for Modeling defocus disparity in dual-pixel sensors result or share you result to me, I could not reproduce your result my mail:...