Hsuan
Hsuan
Hi @AishuaiYao @AishuaiYao , If you want to re-train on ADE20k, you need to load the pre-trained weights of the backbone from PSPNet50. For semantic segmentation tasks, it is really...
Hey @deeptf, The original version of interp layer can shrink the image by just setting the parameter of (s_factor, z_factor) to (0.5, 1). Then it would downsample the original image...
Hey guys, you can refer to #50, there are some advices which can help you to solve the problems!
Do you guys load the pre-trained model? I think you should do that, or the network might learned nothing. If you trained from scratch, you should teach the model to...
Hey @Sixkplus, Thank you for the information, I have tested the mIoU using the official cityscapes script and translate the trainId into regular ID. I got the same result as...
I have found the problem! Finally got the desired results,  It seems that there are some shift in the image if you use `inference.py` to generate the output image....
@Sixkplus , thank you a lot!! I will update the bug codes in recent days >_
And also, the model definition is also a little bit different, you can check it in `model.py`.
Can you tell me the resolution of your input images? 1024*2048 size image give me 0.04 sec on GTX1080. Yes, you can change the code to feed batch of images...
Oh, my graphic card is gtx 1080 ti, but I don't think gtx 1080 will 4 times slower than it. Can you try with single input image with following code?...