Slice RII

This patch would be a Render In Image type patch that would be able to configure multiple outputs for a scene (through settings or an input port), each with corresponding adjustable near and far z clip input parameters.

With some care on the part of the user, this would allow one to construct a scene inside of a render in image that could be output to a stack of sprites, with each one being consecutively blurred more than the last, from front to back, to imitate depth - a pretty common old school animation technique.

It could also be used for other types of volume rendering.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

usefuldesign.au's picture
Re: Slice RII

This seems like an expensive way to generate depth blur. Wouldn't it be more on the money to as for a depth-map output image (or Z-map)?

This would give a pixel for pixel indicator of depth. For each pixel in the 3D scene image a corresponding grey scale value in a separate image indicating the distance from camera. The closer the point in the 3D space to camera the more it the corresponding pixel in the depth-map approaches zero.

The depth image then is used (possibly post-processed by other filters) as a mask to the blur filter — giving the required amount of blur for each pixel.

Speaking of depth image I just happened to be looking at these images tonight: kazuki_takamatsu and google images

gtoledo3's picture
Re: Slice RII

Honestly, don't even worry too much about the depth blur part, that's just a bonus. As far as "expense" idk, I have it working without a patch and it's quick, it just sucks to set it up. There are many other reasons why one would want to take rendering slices (like rendering to actual volumetric media).

I understand traditional methods of doing depth blur with a depth image, etc...

usefuldesign.au's picture
Re: Slice RII

Ok I guess I'm starting to see the potential…

gtoledo3's picture
Re: Slice RII

Thinking about your comment about the depth info, I don't believe that QC can output an appropriate depth channel image with enough resolution that anyone could actually do anything useful.