How can I use semantic confidence for occlusion?

How can I use semantic confidence for occlusion? I want to use the function for occlusion, but there is no explanations in samples and below documentation.

Below link has a little bit information, but I can’t
understand what should I do to use semantic confidence in the situation of suppression “sky” channel.

The video is only added overlay effect on sky channel using Graphics.Blit.
On the video, I didn’t change ARDK example.
What I want to do, showing the AR objects on only the sky channel.

At the moment Semantic confidence is used for shader effects only as it is experimental.

The way the giant yeti in the sky video was made was to render it to a texture and blend that into the scene with the confidence mask.

Add the objects you want to draw to a layer (sky_layer),
Hide it from the main camera.
Add a second camera that renders to texture and set it to only render that layer.

Then in your blit pass you can send in the texture with your cg, and the confidence mask.

In the shader you can then composite the items together.

Thank you for giving me a solution.
I will try it.

I did it. Thank you.

I posted an article on how to use it.