gerwang

Hello, Thanks for your great work! I'm pretty interested in your distance transform loss. But when I want to test the loss separately, I find there is no grad_fn for...

Opt examples run significantly slower than CUDA solver

5

I am using Opt with VS2015 and LLVM 6.0.1 on Windows. When I ran those examples, the Opt method is significantly slower than the corresponding CUDA solver, but the paper...

Shape from shading example does not work

4

Using Opt with LLVM 6.0.1 and CUDA 10.0 on Windows 10. Other examples work fine, but when running shape from shading exmple, OptGN method cannot converge and cost rises. OptLM...

Support running Opt on non default cudaStream

1

I want to launch another CPU thread (call it `CPU thread B`) and use Opt on that thread, and I don't want functions like `Opt_ProblemSolve` to block GPU execution invoked...

for "Face Reconstruction from Voice using Generative Adversarial Networks"

Is it possible to input more than five views?

3

Hi, thanks for the great work! If I want to input more than five views (for example, to fairly compare with works like NeRFactor), is it possible? Should I expect...

Got extremely large face when performing data augmentation

7

I got extremely large face when I implemented the data augmentation method described in the paper. Here is my code ```python import config import numpy as np from lib_dr import...

Because the operator functions are dynamic attributes, when I use the model and convert it to nn.DataParallel, the following error will arise during forwarding: ``` --------------------------------------------------------------------------- RuntimeError Traceback (most recent...

Not using real ChebConv?

1

I noticed there is no self-loop during propagation in `ChebConv_Coma`, therefore the code is not using the "real" laplacian matrix? Is this behavior consistent with the tensorflow version? minimum code...