Dear Community,
I am working on a Projekt where I use the unreal engine as a Image Generator with different light and camera positions for some Maschine learning tasks. To generate the images I used the sceenshot tool and rendertarget/capture2D. I need the rendertarget for generating a shadow mask by subtracing result of one renderpass with and one without the dynamic shadow properiers. Which works very well. Also the Image which is contained after capturing in the rendertarget looks a bit darker in the editor…
But when it come to exporting the rendertarget (via the exoirt rendertarget node) as Png. This image is much darker. I tried setting up the gamma as in very much posts recommanded. but this seems to be different afterall. So i am wondering if they are using different ways to convert the readed buffer or compressing the data in the sceenshot tool?
I would be glad if someone here has a answer to this and can help me out of this.
Snipped Image for Editorview on rendertarget
Image taken from Screenshot function
Image exported via Export node
329855-
Shadow Map extracted via Rendering Diff