-
Dear authors：
I just use vgg16 to train sim2cs_car(sim10k --> cityscapes_car)
At epoch 1 iter 3400/10000 loss equals 0.5670, but I found loss will increase in later iterations to around…
-
Thank you for your awesome code! I've been trying to run some models on Cityscapes including FCN8, PSPNet and DUC, but I haven't been able to replicate the results reported in the papers. So if it's n…
-
I have converted disparity to depth by `depth = baseline * fx / disparity`, and the depth value will be large, maybe about 500. But actually, depth can not be large. In KITTI, it will be limited in 80…
-
I got a error in loss.py line 54. I print my target size is (2,768,768,3), but the code shows it should be dim 3. my predict size is (2,3,768,768).
I use my own dataset, picture annotation like city…
-
After cp rcnn/CXX_OP/* incubator-mxnet/src/operator/ ,then compile,the errors occuring as follow.
src/operator/./roi_align_v1-inl.h:336:9: error: ‘Fill’ was not declared in this scope
Fil…
-
Hi.
First of all, thanks for sharing your great work.
I'd like to reproduce numbers in the paper. However, it seems like the scripts in the repo have slightly different configurations from the pap…
-
Hi,
When I train my model with the default training parameters on the cityscapes data sets I cant seem to produce your quality of images. I've attempted further training but the generator then lear…
-
Hi, Thanks so much for the code!
I see you mentioned in your 'readme' that your Cityscapes Event dataset is generated by EventGAN,
But in the 'demo.py' code of EventGAN, the Event is generated b…
-
从github上下载deeplabv3_xception65_bn的权重参数，直接用paddleseg加载进行预测，使用默认的参数配置，输出的miou只有0.59
[EVAL]#image=500 acc=0.9114 IoU=0.5917
[EVAL]Category IoU: [0.9464 0.6721 0.8490 0.4693 0.4214 0.2727 0.3392 0.476…
-
@strivebo 我之前一直用的isprs，今年下半年跑的cityscapes，
_Originally posted by @niecongchong in https://github.com/niecongchong/DANet-keras/issues/3#issuecomment-568868692_
大佬，你研究了，我也是遥感的，马上研三l，也是用到ispr数据集