Sengupta

While training on real images without supervision, we compose the foreground on the same background but after shifting it left or right. We use a discriminator on the composed image...

np.random.randint(x,y) generates random integers between x and y. This error means y

Segmentation via Deeplab can often fail for these kinds of situations. I am curious about what kind of error the matting network produces. It has been trained to be robust...

It appears that there is a bug in the code which resulted in multi-frame features not being used at all. This was the model also used for producing results in...

The figure 5 in the paper is not valid, but the rest is fine. I just recently did the experiments after correctly using the motion cues and I did not...

The trained model does not use motion cues. However, due to the bug in the networks.py file, you will require to input something as motion cue (Bx4xWxH). You can use...

Due to the bug, it uses background features twice instead of the motion features. `oth_feat=torch.cat([self.comb_back(torch.cat([img_feat,back_feat],dim=1)),self.comb_seg(torch.cat([img_feat,seg_feat],dim=1)),self.comb_multi(torch.cat([img_feat,back_feat],dim=1))],dim=1)` In Fig 5, since we disabled motion features, we actually used background feature only once:...

Just make changes in the code to rename .png to .jpg or you could simply rename .jpg to .png. It should not matter as long as the image can be...

If there is no background image this method won't work.

I think I used 2 GPU and it converges in roughly 30hours.