SideFX Labs
Free and Open-Source Houdini Tools for Everyone
Home
Resume

Labs Physical Ambient Occlusion

The Physical Ambient Occlusion (SOP) node uses updated algorithms to compute physically plausible ambient occlusion, detects valleys or peaks, and generates dirt buildup. It works on points of reasonably high-res meshes. This node consolidates and improves on the functionalities of several older nodes: Mask by Feature (AO mask), Labs Calculate Thickness, and Labs Calculate Occlusion.

Improvements on previous methods: more stable, more physically plausible, additional artistic controls.

Improvements on previous methods: more stable, more physically plausible, additional artistic controls.

The previous algorithm uses a random seed, which does not make too much sense since AO is not a random quality.

The previous algorithm uses a random seed, which does not make too much sense since AO is not a random quality.

The new algorithm has better sample distribution. It is also seedless, so the result is clean and stable.

The new algorithm has better sample distribution. It is also seedless, so the result is clean and stable.

Left: Physical Ambient Occlusion in Houdini with vertex color. Right: Ambient Occlusion render pass in Blender with Cycles.

Left: Physical Ambient Occlusion in Houdini with vertex color. Right: Ambient Occlusion render pass in Blender with Cycles.

Physical Ambient Occlusion includes a Dirt Buildup mode which is the same as Mask by Feature (AO mask) and Calculate Occlusion.

Physical Ambient Occlusion includes a Dirt Buildup mode which is the same as Mask by Feature (AO mask) and Calculate Occlusion.

Physical Ambient Occlusion also includes a Peaks mode which is the same as Calculate Thickness but with a newer algorithm.

Physical Ambient Occlusion also includes a Peaks mode which is the same as Calculate Thickness but with a newer algorithm.

Demonstration of the 4 different modes of this node: Physical Ambient Occlusion, Valleys, Peaks, Dirt Buildup.

Physical Ambient Occlusion uses icosahedral sample distribution, geometry ray tracing, and square root weighting to achieve better physical plausibility. Screenshot from Volumetric Ambient Occlusion [Szirmay-Kalos et al., 2010].

Physical Ambient Occlusion uses icosahedral sample distribution, geometry ray tracing, and square root weighting to achieve better physical plausibility. Screenshot from Volumetric Ambient Occlusion [Szirmay-Kalos et al., 2010].

Report