IвЂ™d presume even though its a black box, the normal controls which pixels вЂњconnectвЂќ for the smearing, so surfaces with viewspace Normal Z>0 would get grouped together with viewspace Normal Z<0 when the magnitude of (length(Z)) is similar.We transmit data to third-party providers in order to improve our web offering for you. These normals are not being used for shading, they are being used in a black-box neural network. So the denoiser was trained using ray-space with -Z being the forward direction of the ray? Or was it camera space?ДЄs far as whether we can or cannot drop the Z component of the normal, indeed the sign ambiguity is known, it just doesnвЂ™t matter. ThatвЂ™s possible and happens in every single kind of perspective projection, only ortho and telelens would not suffer (because they are 100% planar). IвЂ™m pretty sure that would still result in roughly the same denoiser output as camera space normals.

That said, if youвЂ™re denoising and worried about having pixels with +Z normals, you could certainly render some ray-space normals where the -Z direction always points directly at the center of projection, and that way youвЂ™d never get normals with any +Z component. The broader answer is that as long as the processing of normal data used for inference matches the processing of normal data used for training, the neural network will learn to use whatever data it has, and there are no constraints on what channels we use or drop, and no notion of correct or incorrect. Intuitively, you can see that the normals that point along the camera forward vector are all near the edge of an object, so are not very likely to affect the denoiserвЂ™s choices.

If you follow that loosely, the denoiser with a normal buffer should work for you.ДЄs far as whether we can or cannot drop the Z component of the normal, indeed the sign ambiguity is known, it just doesnвЂ™t matter. I will request this section be reviewed.ДЇor the denoiser, the important part is that the normals you use roughly match the training set, which means that you should transform your world space normals so that Z is along the view direction, X is right, and Y is up. Screen space is generally referring to a 2d space, so it doesnвЂ™t usually mean the same thing as either camera space or view space, but the OptiX documentation authors may be referring to camera space without the Z coordinate. You can use any camera projection you want, linear or non-linear, so вЂњcamera spaceвЂќ is the more appropriate choice here, and it happens to be common in ray tracing circles, e.g. But we donвЂ™t want to use вЂњview spaceвЂќ for OptiX, because OptiX is not a raster engine and the camera weвЂ™re talking about is not necessarily a linear-perspective transform. The names of various spaces are overloaded, so view space does technically work, if youвЂ™re talking about OpenGLвЂ™s definition. YouвЂ™re right that we should define those terms, and that it may technically be possible to see normals with ambiguous Z sign, depending on your choice of projection. Still one would think that it would be common knowledge that one cannot drop the Z-component of a normal, even in view space, because thanks to perspective cameras we can see normals that have a positive Z component (pointing towards camera forward vector).

Two different terms are being used вЂњcamera spaceвЂќ and вЂњscreen spaceвЂќ which I assume both mean вЂњview spaceвЂќ (thereвЂ™s not really much use having projection space normals, right?) It must have the same type and dimensions as the input buffer. An optional fourth channel of this buffer is ignored. The camera space is assumed to be right handed such that the camera is looking down the negative z axis, and the up direction is along the y axis. However, because the normal vectors are defined in screen space, the z component is ignored. These two formats can represent the x, y and z components of a vector. The buffer contains data in format OPTIX_PIXEL_FORMAT_HALF3 or OPTIX_PIXEL_FORMAT_FLOAT3. The optional, noise-free normal buffer (OPTIX_DENOISER_INPUT_RGB_ALBEDO_NORMAL) contains the surface normals of the primary hit in camera space. The Programming Guide for the AI Denoiser

Denoiser 3 makes screen black