Florian Bruggisser

Oh, thanks for that simple solution ✨ I guess the wgpu conversion has not been applied in your rendering. It's seems ok now:

@almarklein Would you be open to a PR that adds a compatibility method to load open3d geometries and materials like the one for trimesh? Or do you want to keep...

Same issue about nanodet/model/arch/one_stage_detector.py. No idea why this fails now and didn't before. I am only changing the README.md.

I experienced the same problem with my custom `yolo` dataset reader (#487) which is based on the `xml` reader you are using. It seems that there is a bug in...

I am not sure how VOC XML numbers the categories, but Yolo starts at `0` and I guess MS COCO starts at `1`, that's why my problem was fixed with...

@wjakob Wow, thanks for all the work you put into this extension 🎉. I will archive my project and add a link to the official implementation as soon as it...

I was able to build version 2.54.2 and run the test scripts using `sudo` to display RGB and depth streams. The current framework works, not stable (sometimes you have to...

It should be simple, if we could be able to call this method directly from python. ```c EncodedObject encode_mesh(const std::vector &points, const std::vector &faces, int quantization_bits, int compression_level, float quantization_range,...

Currently this is meant to playback a sequence of already exiting meshes, not like `imshow` by opencv. But it would be possible I guess, here an example which displays a...

Not as it is at the moment, but of course open3d would support to load them. It should be simple to replace the mesh part with the pointcloud loading, but...