Florian Bruggisser

And did you have a look at https://github.com/cansik/librealsense-java/blob/master/src/test/java/org/intel/rs/ui/ProcessingBlockTest.java already?

I think your terminology is a bit misleading. You always work on the `depth-frame`, which by default is Y16 (`16bit unsigned integer`) on a realsense camera. If you map the...

Do you mean just one depth frame or a stream of data (rosbag)? If you just need to open in it in python afterwards, you are also able to just...

Hi I am not sure which version supports the D457, but as it is based on the D450, I think it should work work out of the box with the...

There should be no need to install multiple versions of onnxruntime. Either you install the `onnxruntime-silicon` or `onnxruntime-coreml`. After correct installation, try to check what kind of providers are available:...

Could it be because of the supported [ops](https://onnxruntime.ai/docs/execution-providers/CoreML-ExecutionProvider.html#supported-ops) for the CoreML backend? If they are not implemented, the CPU is used, which can lead to a fragmentation of the calculation...

@heliumsoft Instead of writing into a random issue, please read the [FAQ](https://github.com/cansik/onnxruntime-silicon?tab=readme-ov-file#faq) first. If there are still questions, open an issue where you describe the problem, the environment you are...

@angewandte-codinglab Please do not copy/paste what chat-gpt "thinks". Answer questions with your own knowledge and experience.

Thanks for your contribution, it seems to look good. I am still doing some tests, because I couldn't build it yet, but I guess something has changed on my build...

What exactly is the issue you are facing? You won't have a `GPU` provider for MacOSX, only `CoreMLExecutionProvider` which is available in your example output.