Files
2021-06-13 10:28:03 +02:00

402 lines
13 KiB
HLSL

#ifndef UNIVERSAL_SSAO_INCLUDED
#define UNIVERSAL_SSAO_INCLUDED
// Includes
#include "Packages/com.unity.render-pipelines.core/ShaderLibrary/Common.hlsl"
#include "Packages/com.unity.render-pipelines.universal/ShaderLibrary/ShaderVariablesFunctions.hlsl"
#include "Packages/com.unity.render-pipelines.universal/ShaderLibrary/DeclareDepthTexture.hlsl"
#include "Packages/com.unity.render-pipelines.universal/ShaderLibrary/DeclareNormalsTexture.hlsl"
// Textures & Samplers
TEXTURE2D_X(_BaseMap);
TEXTURE2D_X(_ScreenSpaceOcclusionTexture);
SAMPLER(sampler_BaseMap);
SAMPLER(sampler_ScreenSpaceOcclusionTexture);
// Params
float4 _BlurOffset;
float4 _SSAOParams;
float4 _SourceSize;
// SSAO Settings
#define INTENSITY _SSAOParams.x
#define RADIUS _SSAOParams.y
#define DOWNSAMPLE _SSAOParams.z
// GLES2: In many cases, dynamic looping is not supported.
#if defined(SHADER_API_GLES) && !defined(SHADER_API_GLES3)
#define SAMPLE_COUNT 3
#else
#define SAMPLE_COUNT _SSAOParams.w
#endif
// Function defines
#define SCREEN_PARAMS GetScaledScreenParams()
#define SAMPLE_BASEMAP(uv) SAMPLE_TEXTURE2D_X(_BaseMap, sampler_BaseMap, UnityStereoTransformScreenSpaceTex(uv));
#define SAMPLE_BASEMAP_R(uv) SAMPLE_TEXTURE2D_X(_BaseMap, sampler_BaseMap, UnityStereoTransformScreenSpaceTex(uv)).r;
// Constants
// kContrast determines the contrast of occlusion. This allows users to control over/under
// occlusion. At the moment, this is not exposed to the editor because it's rarely useful.
static const float kContrast = 0.6;
// The constant below controls the geometry-awareness of the bilateral
// filter. The higher value, the more sensitive it is.
static const float kGeometryCoeff = 0.8;
// The constants below are used in the AO estimator. Beta is mainly used for suppressing
// self-shadowing noise, and Epsilon is used to prevent calculation underflow. See the
// paper (Morgan 2011 http://goo.gl/2iz3P) for further details of these constants.
static const float kBeta = 0.002;
#define EPSILON 1.0e-4
float4 PackAONormal(float ao, float3 n)
{
return float4(ao, n * 0.5 + 0.5);
}
float3 GetPackedNormal(float4 p)
{
return p.gba * 2.0 - 1.0;
}
float GetPackedAO(float4 p)
{
return p.r;
}
float EncodeAO(float x)
{
#if UNITY_COLORSPACE_GAMMA
return 1.0 - max(LinearToSRGB(1.0 - saturate(x)), 0.0);
#else
return x;
#endif
}
float CompareNormal(float3 d1, float3 d2)
{
return smoothstep(kGeometryCoeff, 1.0, dot(d1, d2));
}
float2 GetScreenSpacePosition(float2 uv)
{
return uv * SCREEN_PARAMS.xy * DOWNSAMPLE;
}
// Trigonometric function utility
float2 CosSin(float theta)
{
float sn, cs;
sincos(theta, sn, cs);
return float2(cs, sn);
}
// Pseudo random number generator with 2D coordinates
float UVRandom(float u, float v)
{
float f = dot(float2(12.9898, 78.233), float2(u, v));
return frac(43758.5453 * sin(f));
}
// Sample point picker
float3 PickSamplePoint(float2 uv, float randAddon, int index)
{
float2 positionSS = GetScreenSpacePosition(uv);
float gn = InterleavedGradientNoise(positionSS, index);
float u = frac(UVRandom(0.0, index + randAddon) + gn) * 2.0 - 1.0;
float theta = (UVRandom(1.0, index + randAddon) + gn) * TWO_PI;
return float3(CosSin(theta) * sqrt(1.0 - u * u), u);
}
float RawToLinearDepth(float rawDepth)
{
#if defined(_ORTHOGRAPHIC)
#if UNITY_REVERSED_Z
return ((_ProjectionParams.z - _ProjectionParams.y) * (1.0 - rawDepth) + _ProjectionParams.y);
#else
return ((_ProjectionParams.z - _ProjectionParams.y) * (rawDepth) + _ProjectionParams.y);
#endif
#else
return LinearEyeDepth(rawDepth, _ZBufferParams);
#endif
}
float SampleAndGetLinearDepth(float2 uv)
{
float rawDepth = SampleSceneDepth(uv.xy).r;
return RawToLinearDepth(rawDepth);
}
float3 ReconstructViewPos(float2 uv, float depth, float2 p11_22, float2 p13_31)
{
#if defined(_ORTHOGRAPHIC)
float3 viewPos = float3(((uv.xy * 2.0 - 1.0 - p13_31) * p11_22), depth);
#else
float3 viewPos = float3(depth * ((uv.xy * 2.0 - 1.0 - p13_31) * p11_22), depth);
#endif
return viewPos;
}
// Try reconstructing normal accurately from depth buffer.
// Low: DDX/DDY on the current pixel
// Medium: 3 taps on each direction | x | * | y |
// High: 5 taps on each direction: | z | x | * | y | w |
// https://atyuwen.github.io/posts/normal-reconstruction/
// https://wickedengine.net/2019/09/22/improved-normal-reconstruction-from-depth/
float3 ReconstructNormal(float2 uv, float depth, float3 vpos, float2 p11_22, float2 p13_31)
{
#if defined(_RECONSTRUCT_NORMAL_LOW)
return normalize(cross(ddy(vpos), ddx(vpos)));
#else
float2 delta = _SourceSize.zw * 2.0;
// Sample the neighbour fragments
float2 lUV = float2(-delta.x, 0.0);
float2 rUV = float2( delta.x, 0.0);
float2 uUV = float2(0.0, delta.y);
float2 dUV = float2(0.0, -delta.y);
float3 l1 = float3(uv + lUV, 0.0); l1.z = SampleAndGetLinearDepth(l1.xy); // Left1
float3 r1 = float3(uv + rUV, 0.0); r1.z = SampleAndGetLinearDepth(r1.xy); // Right1
float3 u1 = float3(uv + uUV, 0.0); u1.z = SampleAndGetLinearDepth(u1.xy); // Up1
float3 d1 = float3(uv + dUV, 0.0); d1.z = SampleAndGetLinearDepth(d1.xy); // Down1
// Determine the closest horizontal and vertical pixels...
// horizontal: left = 0.0 right = 1.0
// vertical : down = 0.0 up = 1.0
#if defined(_RECONSTRUCT_NORMAL_MEDIUM)
uint closest_horizontal = l1.z > r1.z ? 0 : 1;
uint closest_vertical = d1.z > u1.z ? 0 : 1;
#else
float3 l2 = float3(uv + lUV * 2.0, 0.0); l2.z = SampleAndGetLinearDepth(l2.xy); // Left2
float3 r2 = float3(uv + rUV * 2.0, 0.0); r2.z = SampleAndGetLinearDepth(r2.xy); // Right2
float3 u2 = float3(uv + uUV * 2.0, 0.0); u2.z = SampleAndGetLinearDepth(u2.xy); // Up2
float3 d2 = float3(uv + dUV * 2.0, 0.0); d2.z = SampleAndGetLinearDepth(d2.xy); // Down2
const uint closest_horizontal = abs( (2.0 * l1.z - l2.z) - depth) < abs( (2.0 * r1.z - r2.z) - depth) ? 0 : 1;
const uint closest_vertical = abs( (2.0 * d1.z - d2.z) - depth) < abs( (2.0 * u1.z - u2.z) - depth) ? 0 : 1;
#endif
// Calculate the triangle, in a counter-clockwize order, to
// use based on the closest horizontal and vertical depths.
// h == 0.0 && v == 0.0: p1 = left, p2 = down
// h == 1.0 && v == 0.0: p1 = down, p2 = right
// h == 1.0 && v == 1.0: p1 = right, p2 = up
// h == 0.0 && v == 1.0: p1 = up, p2 = left
// Calculate the view space positions for the three points...
float3 P1;
float3 P2;
if (closest_vertical == 0)
{
P1 = closest_horizontal == 0 ? l1 : d1;
P2 = closest_horizontal == 0 ? d1 : r1;
}
else
{
P1 = closest_horizontal == 0 ? u1 : r1;
P2 = closest_horizontal == 0 ? l1 : u1;
}
P1 = ReconstructViewPos(P1.xy, P1.z, p11_22, p13_31);
P2 = ReconstructViewPos(P2.xy, P2.z, p11_22, p13_31);
// Use the cross product to calculate the normal...
return normalize(cross(P2 - vpos, P1 - vpos));
#endif
}
void SampleDepthNormalView(float2 uv, float2 p11_22, float2 p13_31, out float depth, out float3 normal, out float3 vpos)
{
depth = SampleAndGetLinearDepth(uv);
vpos = ReconstructViewPos(uv, depth, p11_22, p13_31);
#if defined(_SOURCE_DEPTH_NORMALS)
normal = SampleSceneNormals(uv);
#else
normal = ReconstructNormal(uv, depth, vpos, p11_22, p13_31);
#endif
}
float3x3 GetCoordinateConversionParameters(out float2 p11_22, out float2 p13_31)
{
float3x3 camProj = (float3x3)unity_CameraProjection;
p11_22 = rcp(float2(camProj._11, camProj._22));
p13_31 = float2(camProj._13, camProj._23);
return camProj;
}
// Distance-based AO estimator based on Morgan 2011
// "Alchemy screen-space ambient obscurance algorithm"
// http://graphics.cs.williams.edu/papers/AlchemyHPG11/
float4 SSAO(Varyings input) : SV_Target
{
UNITY_SETUP_STEREO_EYE_INDEX_POST_VERTEX(input);
float2 uv = input.uv;
// Parameters used in coordinate conversion
float2 p11_22, p13_31;
float3x3 camProj = GetCoordinateConversionParameters(p11_22, p13_31);
// Get the depth, normal and view position for this fragment
float depth_o;
float3 norm_o;
float3 vpos_o;
SampleDepthNormalView(uv, p11_22, p13_31, depth_o, norm_o, vpos_o);
// This was added to avoid a NVIDIA driver issue.
float randAddon = uv.x * 1e-10;
float rcpSampleCount = rcp(SAMPLE_COUNT);
float ao = 0.0;
for (int s = 0; s < int(SAMPLE_COUNT); s++)
{
#if defined(SHADER_API_D3D11)
// This 'floor(1.0001 * s)' operation is needed to avoid a DX11 NVidia shader issue.
s = floor(1.0001 * s);
#endif
// Sample point
float3 v_s1 = PickSamplePoint(uv, randAddon, s);
// Make it distributed between [0, _Radius]
v_s1 *= sqrt((s + 1.0) * rcpSampleCount ) * RADIUS;
v_s1 = faceforward(v_s1, -norm_o, v_s1);
float3 vpos_s1 = vpos_o + v_s1;
// Reproject the sample point
float3 spos_s1 = mul(camProj, vpos_s1);
#if defined(_ORTHOGRAPHIC)
float2 uv_s1_01 = clamp((spos_s1.xy + 1.0) * 0.5, 0.0, 1.0);
#else
float2 uv_s1_01 = clamp((spos_s1.xy * rcp(vpos_s1.z) + 1.0) * 0.5, 0.0, 1.0);
#endif
// Depth at the sample point
float depth_s1 = SampleAndGetLinearDepth(uv_s1_01);
// Relative position of the sample point
float3 vpos_s2 = ReconstructViewPos(uv_s1_01, depth_s1, p11_22, p13_31);
float3 v_s2 = vpos_s2 - vpos_o;
// Estimate the obscurance value
float a1 = max(dot(v_s2, norm_o) - kBeta * depth_o, 0.0);
float a2 = dot(v_s2, v_s2) + EPSILON;
ao += a1 * rcp(a2);
}
// Intensity normalization
ao *= RADIUS;
// Apply contrast
ao = PositivePow(ao * INTENSITY * rcpSampleCount, kContrast);
return PackAONormal(ao, norm_o);
}
// Geometry-aware separable bilateral filter
half4 Blur(float2 uv, float2 delta) : SV_Target
{
float4 p0 = SAMPLE_BASEMAP(uv );
float4 p1a = SAMPLE_BASEMAP(uv - delta * 1.3846153846);
float4 p1b = SAMPLE_BASEMAP(uv + delta * 1.3846153846);
float4 p2a = SAMPLE_BASEMAP(uv - delta * 3.2307692308);
float4 p2b = SAMPLE_BASEMAP(uv + delta * 3.2307692308);
#if defined(BLUR_SAMPLE_CENTER_NORMAL)
#if defined(_SOURCE_DEPTH_NORMALS)
float3 n0 = SampleSceneNormals(uv);
#else
float2 p11_22, p13_31;
float3x3 camProj = GetCoordinateConversionParameters(p11_22, p13_31);
// Get the depth, normal and view position for this fragment
float depth_o;
float3 n0;
float3 vpos_o;
SampleDepthNormalView(uv, p11_22, p13_31, depth_o, n0, vpos_o);
#endif
#else
float3 n0 = GetPackedNormal(p0);
#endif
float w0 = 0.2270270270;
float w1a = CompareNormal(n0, GetPackedNormal(p1a)) * 0.3162162162;
float w1b = CompareNormal(n0, GetPackedNormal(p1b)) * 0.3162162162;
float w2a = CompareNormal(n0, GetPackedNormal(p2a)) * 0.0702702703;
float w2b = CompareNormal(n0, GetPackedNormal(p2b)) * 0.0702702703;
float s;
s = GetPackedAO(p0) * w0;
s += GetPackedAO(p1a) * w1a;
s += GetPackedAO(p1b) * w1b;
s += GetPackedAO(p2a) * w2a;
s += GetPackedAO(p2b) * w2b;
s *= rcp(w0 + w1a + w1b + w2a + w2b);
return PackAONormal(s, n0);
}
// Geometry-aware bilateral filter (single pass/small kernel)
float BlurSmall(float2 uv, float2 delta)
{
float4 p0 = SAMPLE_BASEMAP(uv );
float4 p1 = SAMPLE_BASEMAP(uv + float2(-delta.x, -delta.y));
float4 p2 = SAMPLE_BASEMAP(uv + float2( delta.x, -delta.y));
float4 p3 = SAMPLE_BASEMAP(uv + float2(-delta.x, delta.y));
float4 p4 = SAMPLE_BASEMAP(uv + float2( delta.x, delta.y));
float3 n0 = GetPackedNormal(p0);
float w0 = 1.0;
float w1 = CompareNormal(n0, GetPackedNormal(p1));
float w2 = CompareNormal(n0, GetPackedNormal(p2));
float w3 = CompareNormal(n0, GetPackedNormal(p3));
float w4 = CompareNormal(n0, GetPackedNormal(p4));
float s;
s = GetPackedAO(p0) * w0;
s += GetPackedAO(p1) * w1;
s += GetPackedAO(p2) * w2;
s += GetPackedAO(p3) * w3;
s += GetPackedAO(p4) * w4;
return s *= rcp(w0 + w1 + w2 + w3 + w4);
}
half4 HorizontalBlur(Varyings input) : SV_Target
{
UNITY_SETUP_STEREO_EYE_INDEX_POST_VERTEX(input);
float2 uv = input.uv;
float2 delta = float2(_SourceSize.z * rcp(DOWNSAMPLE) * 2.0, 0.0);
return Blur(uv, delta);
}
half4 VerticalBlur(Varyings input) : SV_Target
{
UNITY_SETUP_STEREO_EYE_INDEX_POST_VERTEX(input);
float2 uv = input.uv;
float2 delta = float2(0.0, _SourceSize.w * rcp(DOWNSAMPLE) * 2.0);
return Blur(uv, delta);
}
half4 FinalBlur(Varyings input) : SV_Target
{
UNITY_SETUP_STEREO_EYE_INDEX_POST_VERTEX(input);
float2 uv = input.uv;
float2 delta = _SourceSize.zw * rcp(DOWNSAMPLE);
return 1.0 - BlurSmall(uv, delta );
}
#endif //UNIVERSAL_SSAO_INCLUDED