Sagnik Das

Results 13 comments of Sagnik Das

We were having data storage issues. It is solved now. I will send the credentials soon. Please try downloading after you receive the uname and passw. Apologies for the inconvenience!

yeah, input is the image, load the wc ground-truth, and bm ground-truth.

Sorry for delaying this, will try to do this by the weekend. You can also do it pretty easily, just combine two training codes and rename the variables accordingly. In...

I believe this is happening due to some error in the shape network regression, it could be related to normalization. Could you please share the output of the shape network?

This issue is resolved if you use scipy (

1. Yes, that's how shading map is calculated. Make sure to scale the values of the map. 2. The normal estimation network is pretrained with L2. While training Shading estimation...

Do something like this: `shdmap=np.divide(img,alb,out=np.zeros_like(img), where=alb!=0).astype(np.float32)`

We didn't scale the values. Used ReLU as the final activation for shading map regression. However, people often use log scale.

You have to wait for a week or so. We are trying to figure out a better way to serve the dataset. Due to high demands, we're currently experiencing server...