Erika Lu
Erika Lu
Hi, this result looks reasonable to me. The shape of the head is a bit strange, but our layered neural rendering model can correct for slight issues like that, so...
Oh the pink is just the body part ID. The RGB channels correspond to IUV, where I is an integer 1-25 ( *10 for visualization purposes). The losses are the...
The matching_net is a network that takes as input the concatenated features for the exemplar patch and image. So the matching_net can therefore be used to compare the exemplar patch...
Hi, can you try resizing the patch to 64x64? The adapted model actually uses a different patch size, sorry for the confusion. To answer your question about the dot annotations,...
If you use the default config settings (in dataloader.py) to adapt to the VGG cell dataset, you will have to modify the demo script to make sure the inputs match...
Can you right-click on that link and save it?
@mcahny A demo script would be really helpful. Is there any update on when this might happen?
Hi, good catch, the regularization should be applied to the alpha composite, as it is in the code and not the paper. This reason for this is to allow for...
The method we used to compute homographies was more accurate for higher resolution images, which is why we added the option to use homographies computed on a different resolution. You...
Hi, you can run the demo in the RAFT repo: https://github.com/princeton-vl/RAFT Run it on the directory of frames to get the forward flow, and to get the backward flow just...