Merge pull request #20 from occ-ai/roy.pixelate_dilate_zoom_choices

feat: Add pixelate effect support for masking

README.md

@@ -39,18 +39,18 @@ Current features:
 
 - Detect over 80 categories of objects, using an efficient model ([EdgeYOLO](https://github.com/LSH9832/edgeyolo))
 - 3 Model sizes: Small, Medium and Large
+- Face detection model, fast and efficient ([YuNet](https://github.com/opencv/opencv_zoo/tree/main/models/face_detection_yunet))
 - Load custom ONNX detection models from disk
-- Control detection threshold
-- Select object category filter (e.g. find only "Person")
-- Masking: Blur, Solid color, Transparent, output binary mask (combine with other plugins!)
-- Tracking: Single object / All objects, Zoom factor, smooth transition
+- Filter by: Minimal Detection confidence, Object category (e.g. only "Person"), Object Minimal Size
+- Masking: Blur, Pixelate, Solid color, Transparent, output binary mask (combine with other plugins!)
+- Tracking: Single object / Biggest / Oldest / All objects, Zoom factor, smooth transition
+- SORT algorithm for tracking smoothness and continuity
+- Save detections to file in real-time, for integrations e.g. with Streamer.bot
 
 Roadmap features:
 - Precise object mask, beyond bounding box
-- Implement SORT tracking for smoothness
 - Multiple object category selection (e.g. Dog + Cat + Duck)
 - Make available detection information for other plugins through settings
-- More real-time models choices
 
 ## Train and use a custom detection model
 

data/effects/pixelate.effect

@@ -0,0 +1,51 @@
+uniform float4x4 ViewProj;
+uniform texture2d image;
+uniform texture2d focalmask;
+
+uniform float pixel_size; // Size of the pixelation
+uniform float2 tex_size; // Size of the texture in pixels
+
+sampler_state textureSampler {
+	Filter    = Linear;
+	AddressU  = Clamp;
+	AddressV  = Clamp;
+};
+
+struct VertDataIn {
+	float4 pos : POSITION;
+	float2 uv  : TEXCOORD0;
+};
+
+struct VertDataOut {
+	float4 pos : POSITION;
+	float2 uv  : TEXCOORD0;
+};
+
+VertDataOut VSDefault(VertDataOut v_in)
+{
+	VertDataOut vert_out;
+	vert_out.pos = mul(float4(v_in.pos.xyz, 1.0), ViewProj);
+	vert_out.uv  = v_in.uv;
+	return vert_out;
+}
+
+float4 PSPixelate(VertDataOut v_in) : TARGET
+{
+    if (focalmask.Sample(textureSampler, v_in.uv).r == 0) {
+		// No mask - return the original image value without any blur
+		return image.Sample(textureSampler, v_in.uv);
+	}
+
+    float2 pixelUV = v_in.uv * tex_size; // Convert to pixel coordinates
+    float2 pixelatedUV = floor(pixelUV / pixel_size) * pixel_size / tex_size;
+    return image.Sample(textureSampler, pixelatedUV);
+}
+
+technique Draw
+{
+	pass
+	{
+		vertex_shader = VSDefault(v_in);
+		pixel_shader  = PSPixelate(v_in);
+	}
+}

data/locale/en-US.ini

@@ -7,7 +7,7 @@ GPUTensorRT="GPU (TensorRT)"
 GPUDirectML="GPU (DirectML)"
 CoreML="CoreML"
 NumThreads="Number of Threads"
-ModelSize="Model Size"
+ModelSize="Model"
 SmallFast="Small (Fast)"
 Medium="Medium"
 LargeSlow="Large (Accurate)"
@@ -22,7 +22,7 @@ Blur="Blur"
 OutputMask="Output Mask"
 Transparent="Transparent"
 MaskingColor="Masking Color"
-MaskingBlurRadius="Masking Blur Radius"
+MaskingBlurRadius="Blur / Pixelate Size"
 TrackingZoomFollowGroup="Tracking (Zoom, Follow) Options"
 ZoomFactor="Zoom Factor"
 ZoomObject="Zoom Object"
@@ -40,3 +40,9 @@ CropLeft="Left"
 CropTop="Top"
 CropRight="Right"
 CropBottom="Bottom"
+Pixelate="Pixelate"
+DilationIterations="Dilation"
+Biggest="Biggest"
+Oldest="Oldest"
+FaceDetect="Face Detection"
+MinSizeThreshold="Min. Object Area"

src/FilterData.h

@@ -17,11 +17,13 @@ struct filter_data {
 	float conf_threshold;
 	std::string modelSize;
 
+	int minAreaThreshold;
 	int objectCategory;
 	bool maskingEnabled;
 	std::string maskingType;
 	int maskingColor;
 	int maskingBlurRadius;
+	int maskingDilateIterations;
 	bool trackingEnabled;
 	float zoomFactor;
 	float zoomSpeedFactor;
@@ -46,6 +48,7 @@ struct filter_data {
 	gs_stagesurf_t *stagesurface;
 	gs_effect_t *kawaseBlurEffect;
 	gs_effect_t *maskingEffect;
+	gs_effect_t *pixelateEffect;
 
 	cv::Mat inputBGRA;
 	cv::Mat outputPreviewBGRA;

src/consts.h

@@ -9,6 +9,7 @@ const char *const USEGPU_COREML = "coreml";
 
 const char *const KAWASE_BLUR_EFFECT_PATH = "effects/kawase_blur.effect";
 const char *const MASKING_EFFECT_PATH = "effects/masking.effect";
+const char *const PIXELATE_EFFECT_PATH = "effects/pixelate.effect";
 
 const char *const PLUGIN_INFO_TEMPLATE =
 	"<a href=\"https://github.com/occ-ai/obs-detect/\">Detect Plugin</a> (%1) by "