Matt Li

## Explanation ```python from open3d.visualization.tensorboard_plugin import summary ``` This line will run the `MonkeyPatching` method at below https://github.com/isl-org/Open3D/blob/c3dfcbdc8d7cdf5603367b783b6d4a43fc934bd5/python/open3d/visualization/tensorboard_plugin/summary.py#L804-L810 It's really a kind of trick. Be careful, most IDE will remove...

@einarf We write it with our own code, but `meshlab` could render it correctly

```console pytorch 1.9.1 py3.9_cuda11.1_cudnn8.0.5_0 pytorch pytorch3d 0.6.0 pypi_0 pypi ``` ```python RasterizationSettings(image_size=(h, w), blur_radius=0, faces_per_pixel=1, perspective_correct=False) # with PerspectiveCameras() ``` This issue still there by ```python (render_images != render_images).sum() ```...

Use `trimesh` as a temporary alternative. ```python import trimesh trimesh.load(obj_file) colors = torch.Tensor(tri_mesh.visual.vertex_colors) # [N, C], Channel = 4 for default. ``` Btw, `pytorch3d` can not load files by specific...

@Caenorst Thank you for responding. But this data is `usd` format, I'm trying to find a way to convert it to an `obj` file, do you have any suggestions?

@Caenorst Due to the hardware requirement of kaolin, I can not run it on my computer with only one 1060ti GPU. I just found a way to load the `.usd`...

@Crghilardi Seems ok, I replace `/Werror to /W4`, and remove `/Wextra` manually. I use cmake to build vs solution, and not find where the code is. Here is what I...

> Hi, we follow the experimental settings of [EdgeConnect](https://arxiv.org/pdf/1901.00212.pdf). For CelebA, we cropped the center 178x178 of the images, then resized them to 256x256 using bilinear interpolation. Thanks a lot,...

> > Hi, maybe it is caused by the preprocessing of the image. We follow the experimental settings of [EdgeConnect](https://arxiv.org/pdf/1901.00212.pdf). For CelebA, we cropped the center 178x178 of the images,...

Make a [PR](https://github.com/aboSamoor/polyglot/pull/234) already