diff --git a/filter/smartDeNoise.glsl b/filter/smartDeNoise.glsl
new file mode 100644
index 00000000..c8607e3a
--- /dev/null
+++ b/filter/smartDeNoise.glsl
@@ -0,0 +1,70 @@
+/*
+original_author: Michele Morrone
+description: |
+    Fast glsl spatial deNoise filter, with circular gaussian kernel and smart/flexible/adaptable -> full configurable
+    * Standard Deviation sigma radius
+    * K factor sigma coefficient
+    * Edge sharpening threshold
+    More about it at https://github.com/BrutPitt/glslSmartDeNoise/
+
+use: <vec4> smartDeNoise(<sampler2D> tex, <vec2> uv, <vec2> pixel, <float> sigma, <float> kSigma, <float> threshold)
+options:
+    - SMARTDENOISE_TYPE
+    - SMARTDENOISE_SAMPLER_FNC(TEX, UV)
+    - SAMPLER_FNC(TEX, UV)
+license: |
+    BSD 2-Clause License
+    Copyright (c) 2018-2020 Michele Morrone
+    All rights reserved.
+*/
+
+#include "../sample.glsl"
+#include "../math/const.glsl"
+
+#ifndef SMARTDENOISE_TYPE
+#define SMARTDENOISE_TYPE vec4
+#endif
+
+#ifndef SMARTDENOISE_SAMPLER_FNC
+#define SMARTDENOISE_SAMPLER_FNC(TEX, UV) SAMPLER_FNC(TEX, UV)
+#endif
+
+#ifndef FNC_SMARTDENOISE
+#define FNC_SMARTDENOISE
+
+SMARTDENOISE_TYPE smartDeNoise(SAMPLER_TYPE tex, vec2 uv, vec2 pixel, float sigma, float kSigma, float threshold) {
+    float radius = floor(kSigma*sigma + 0.5);
+    float radQ = radius * radius;
+    
+    float invSigmaQx2 = 0.5 / (sigma * sigma);      // 1.0 / (sigma^2 * 2.0)
+    float invSigmaQx2PI = INV_PI * invSigmaQx2;    // 1.0 / (sqrt(PI) * sigma)
+    
+    float invThresholdSqx2 = 0.5 / (threshold * threshold);  // 1.0 / (sigma^2 * 2.0)
+    float invThresholdSqrt2PI = INV_SQRT_TAU / threshold;   // 1.0 / (sqrt(2*PI) * sigma)
+    
+    SMARTDENOISE_TYPE centrPx = SMARTDENOISE_SAMPLER_FNC(tex, uv);
+    
+    float zBuff = 0.0;
+    SMARTDENOISE_TYPE aBuff = SMARTDENOISE_TYPE(0.0);
+    for(float x=-radius; x <= radius; x++) {
+        // circular kernel
+        float pt = sqrt(radQ-x*x);
+        for(float y=-pt; y <= pt; y++) {
+            vec2 d = vec2(x,y);
+
+            // gaussian factor
+            float blurFactor = exp( -dot(d , d) * invSigmaQx2 ) * invSigmaQx2PI; 
+            SMARTDENOISE_TYPE walkPx = SMARTDENOISE_SAMPLER_FNC(tex,uv+d*pixel);
+
+            // adaptive
+            SMARTDENOISE_TYPE dC = walkPx-centrPx;
+            float deltaFactor = exp( -dot(dC, dC) * invThresholdSqx2) * invThresholdSqrt2PI * blurFactor;
+                                 
+            zBuff += deltaFactor;
+            aBuff += deltaFactor*walkPx;
+        }
+    }
+    return aBuff/zBuff;
+}
+
+#endif
\ No newline at end of file
diff --git a/math/atan2.glsl b/math/atan2.glsl
new file mode 100644
index 00000000..d56facbc
--- /dev/null
+++ b/math/atan2.glsl
@@ -0,0 +1,15 @@
+#include "const.glsl"
+
+/*
+original_author: Alexander Griffis
+description: |
+    The range here in degrees is 0 to pi (0-180 degrees) and -pi to 0 (181 to 359 degrees
+    More about it at https://github.com/Yaazarai/GLSL-ATAN2-DOC
+
+use: <float> atan2(<float>y, <float> x)
+*/
+
+#ifndef FNC_ATAN2
+#define FNC_ATAN2
+float atan2(float y, float x) { return mod(atan(y,x) + PI, TAU); }
+#endif
\ No newline at end of file