Cecilia

> did you used the syncd bn option？ I don't think I used it cause I just have one gpu to train it.

@sfjuocekr Hello, may I ask where can I find the tutorial about sending depth maps together with rgb images into Metashape to 3d reconstruction??? Thanks a lot!!!

@natowi Anyway, thanks again for your reply, I'll check it out by myself in related tutorial.

朋友，你是不是没有用1-ssim作为loss，我遇到了差不多的问题，发现好像是忘记1-ssim了。

sounds like use the light one as a refine module, which just like the residual refine module, isn't it ?

Wondering about the "fix the heavy u2net", cause the model/train script you offered, the input shape is 288, but when setting the input as 512, can one get good result...

> you can try test time augmentation to get more stable results. As for the convergence, we can't give exact reasons without debugging your code. You may have to double...

> Hi, Recently, we did literature survey on the high resolution image segmentation. We found that combining our U^2-Net with Cascaded PSPNet (https://github.com/hkchengrex/CascadePSP) is a good option for high resolution...