riskeverything

I use the defalut parameters. lr = 1e--5 nums = 15000 in fact, I test in Set14，and get the results as: x2,psnr:29.75,ssim:0.88 however, the original txt is: x2,psnr:33.00,ssim:0.9108. I want...

adaptiveavgpool2d

1

这个应该支持adaptiveavgpool2d的层操作吧？应该怎样做，有了解的嘛？

hi,when I use the Test-Img.py.it told me “cannot save I,16”.so,i change the py. Img =img.astype (‘uint16’) Img =img.convert (‘RGB’) Img.save() it works. But I don’t know why you used disp...

the warp() has some bugs,the dimension of flow and identity_flow is not same,they can't have a "+"?

ir效果对比

3

嗨，你好，最近在复现你们的工作，使用depth图的时候，我的最好效果比文章中只用CASIA数据集还要好一点，但是只使用ir时的效果会很差，这是可以理解的，所以想问下，你们用ir单独训练的效果在验证集上的结果。同时，在训练depth和ir中均出现了严重的过拟合，网络会把召回率降为0，但此时查准率就会变差很多，而且不同模型都是这个趋势，请问，有什么思路解决嘛？

Questions about installing ObjRecRANSAC

1

I installed CUDA in my PC,so it showed "CUDA Found and enabled with OBJREC_USE_CUDA=TRUE, Using CUDA GPU Device ".However,some problems didn't reslove.(1):CMake Error at CMakeLists.txt:16 (cmake_policy):Policy "CMP0053" is not known...