irolaina
Your tensorflow code may have bugs, it is very hard to train.
Hi~ I have trained your network use the code provided and hyper-parameter in the paper. But I find it is very very hard to train, even the training dataset gets the blurry depth prediction. I have also tried fine-tuning your model (.npy file ), the final result is also unsatisfied. I am confused for a long time. I have asked several people, all of them met the same problem I am facing. So can you double check your code, or release your training code. No matter tensorflow version or matlab version.
@luohongcheng hello, I also tried to train the network, but got a blurry depth map too, Do you find some reasons causing it? thank you.
Hi guys, we are currently looking into this. Since our model was trained with matconvnet and then converted to tf for inference we do not know if we get the same performance directly in tf. We will let you know what we find soon. The matlab version should train fine in case you have a setup to try this.
Of course we are open for suggestions how to make the tf code better.
