I have a couple of questions about the best way to implement some features, because I don’t like anything I thought of so far.
In our application we need an image processing functionality: multiple images as an input, each one of them processed conditionally by different shaders and it all combined together as inputs to the main shader that generates end-result. Actual processing steps could be switched or replaced based on configuration parameters in the app.
Sounds very much like a PostProcessPipeline, right? The problem is that postProcess pipeline requires camera and does not work with a texture as an input. Plus I’m not 100% sure how to control when postProcessPipeline actually runs in babylonjs. We don’t need to recalculate everything on every frame.
Another thing that we need to implement is custom mipmap generation as part of the same processing pipeline.
I haven’t found much relevant information for this questions. Could you point me at least in the general direction of possible solutions? Thank you!