Closed strangerattractor closed 4 years ago
It's not a bug. It's intended. There is no such thing like "depth channel" in texture. You can use only RGBA. I use red-channel-only 16 bit texture fo storing data because GBA channels are redundant.
And please do not mix this texture context with "color buffer" and "depth buffer" difference in 3D rendering.
If you want to have black to white visualization using depth information. You can connect red channel to RGB channels so that you can convert "black to red" scale to "black to white" scale which is exactly like the "normal" depth map visualization.
If you want to have black to white visualization using depth information. ...
@chikashimiyama I did that. But why is my picture then showing brighter objects as brighter and not closer objects as brighter?
Do I have to convert it before using it for position?
I put the red channel from your texture out to rgb. The outcome is the above.
It looks like the infrared image, not the depth image.
so what I actually need to work with is the depth image, not the infrared image.
a) RGB b) depth c) infrared
what I need is b) to work with, but right now I get c) correct?
replace infrared with depth frame. plz test.
it works, perfectly. But now I have to check how to properly scale the depth effect, for example in shadergraph.
I have been trying to optimize the ShaderGraph Demo towards proper proportions today.
While working with it I think I found a mistake.
The Texture coming from the Kinect looks like it is actually the red channel, and not the depth channel coming out of the texture.
Am I using the texture wrong or are you giving out the actual red channel and not the depth channel from the kinect?
Or it is kind of a combination of different channels? depth and something?
while searching for examples i found that the depth output texture of the kinect should look more like this![image](https://user-images.githubusercontent.com/52496070/68771204-d37e9180-0627-11ea-8df8-50ebbfdddaa7.png)