NPR技术(2)—Hatching 素描渲染

转载请注明出处：http://blog.csdn.net/tianhai110

 

Hatching（素描）渲染——通过光照的强弱，用不同深浅的纹理，实现一种类似素描效果。

 

 

Hatching的实现原理是：通过N.L计算点受光的强弱，将该值分成不同的段，不同的光照强度对应不同深浅的素描贴图

 

算法对应的VS如下：

VS_OUTPUT vs_main( VS_INPUT Input ) { VS_OUTPUT Output; Output.Position = mul( Input.Position, matViewProjection ); Output.Texcoord = Input.Texcoord; float3 normalW = mul( Input.Normal, matView); float diffuse = min( 1, max(0, dot( vecLightDir, normalW))); //把diffuse平方两次是为了，把亮度降低点，把整体亮度向暗方向偏移，以得到更好的视觉效果。 diffuse = diffuse * diffuse; diffuse = diffuse * diffuse; float factor = diffuse * 6.0; // 通过两个weight,来记录权值，减少传入ps的变量数。 float3 weight0 = 0.0; float3 weight1 = 0.0; // 不同光照强度，对应不同的权值，及不同的纹理贴图 if( factor > 5.0) { weight0.x = 1; } else if( factor > 4.0) { weight0.x = factor - 4.0; weight0.y = 1.0 - weight0.x; } else if( factor > 3.0) { weight0.y = factor - 3.0; weight0.z = 1.0 - weight0.y; } else if( factor > 2.0) { weight0.z = factor - 2.0; weight1.x = 1.0 - weight0.z; } else if( factor > 1.0) { weight1.x = factor - 1.0; weight1.y = 1.0 - weight1.x; } else if( factor > 0.0) { weight1.y = factor; weight1.z = 1.0 - weight1.y; } Output.Weight0 = weight0; Output.Weight1 = weight1; return( Output ); }

PS为：

struct PS_INPUT { float2 tex : TEXCOORD0; float3 weight0 : TEXCOORD1; float3 weight1 : TEXCOORD2; }; float4 ps_main( PS_INPUT input) : COLOR0 { float4 hattch0 = tex2D(hattch0Map, input.tex)*input.weight0.x; float4 hattch1 = tex2D(hattch1Map, input.tex)*input.weight0.y; float4 hattch2 = tex2D(hattch2Map, input.tex)*input.weight0.z; float4 hattch3 = tex2D(hattch3Map, input.tex)*input.weight1.x; float4 hattch4 = tex2D(hattch4Map, input.tex)*input.weight1.y; float4 hattch5 = tex2D(hattch5Map, input.tex)*input.weight1.z; return hattch0 + hattch1 + hattch2 + hattch3 + hattch4 + hattch5; }

 

最终效果