CNHNLP

RefineDet dataArgument

1

Hello @lzx1413 . Do you use the data argument of the SSD paper in the refineDet or SSD?

AIU

4

Anther question； Can you release the evaluation code of AIU?

Question

6

When I evaluate the result on my dataset, I need to process nms and thinned to one pixel wide on GT and crack detection . I run eval/nms_process.m to process...

deconvolution weight

1

Hello,thanks for your code. But there may be some question in deconvolution weight initialize.In original paper,it initialize the deconvolution weight with bilinear interpoltion. And I can't find this in your...

FCN-8s trained to segment Cityscapes

3

hello,I use the pretrained model (FCN-8s trained to segment Cityscapes) to evaluate on the cityscapes val set, and the IOU is very poor (44%). Have you confirmed the pretrained models?

hello,I use the pretrained model (FCN-8s trained to segment Cityscapes) to evaluate on the cityscapes val set, and the IOU is very poor. Do you evaluate the pretrained model? and...

Question

1

When we evaluate the result，we need to run the EdgeEvalDir.m，but what's the meaning of maxDist. I am confused about it and there isn't code.Can you explian it or give some...

原论文中generator最后是Sigmoid，但是代码中使用的是tanh; 可以解释一下吗？ 除此之外，APE-GAN在ImageNet上的结构能够公布一下源代码吗?

The same question with the first issue. Can you give some details about the training process? thank you