Ramana Sundararaman
Ramana Sundararaman
The huber loss is the least squared error if the absolute difference is less than a threshold(which is defined to be 1 for `SmoothL1Loss`), else it is `δ (|d|−δ/2)` where...
Hello, From the `data/generate_data_animals.py` I can generate dogs, cows, cats and hippos. However, I would like to generate other categories such as horses - using the SMALL-R dataset you've mentioned...
Hello Mike, I was wondering if it would be possible to have a mesh w/o orientation if given edges and vertices? i.e, can I create `trimesh.Trimesh` w/o face but edges...
Hello, Congrats on this amazing work and thanks for open-sourcing the code. I am attempting to run the notebook for 3D shape reconstruction locally and I face the error attached...
We can have an example of a Siamese network (similar to Sander Dieleman's [suggestion](https://github.com/Lasagne/Lasagne/issues/168#issuecomment-81134242) ) and convolutional spatial transformer as discussed [here](http://papers.nips.cc/paper/6487-universal-correspondence-network.pdf). From my understanding, the convolutional spatial transformer applies...
Hello, I would like to compute for a point in space it's inside or outside wrt my watertight mesh and then the distance to a face. I see that I...
Hi, Thank you for open-sourcing your code, I am thankful for your time and efforts. I am trying your codebase on a Custom dataset and I have created the JBL...
Hello, Congrats on this amazing work! I'm a little confused between the usage of normal information. From the paper (Eqn. 2) Chamfer's Distance is used only between deformed vertex positions...
Would be great to swap rows and columns in a table
Hi, I was wondering if you'd provide LU/Cholesky/QR factorization for sparse matrices? https://github.com/rgl-epfl/cholespy library provides for cholesky but one would need PD matrices. If I'd want to do LU, I...