Added Yunet face detection example

examples/computer_vision/yunetFaceDetection/Readme.md

@@ -0,0 +1,19 @@
+# Face detection with yunet
+
+### Face detection in OpenFrameworks using dnn
+
+It turns out that doing face detection in 2025 is much more efficient with neural networks than cascades.
+So here is an OF example that uses opencv's facedetectoryn (for yunet).
+
+You will need the file with the pre-trained model. The model version must be compatible with your opencv version.
+
+How to find out your opencv version?
+ * It will be printed out during the ofApp::setup() function.
+ * If you have a version greater than 4.9, you can download the current [face_detection_yunet_2023mar.onnx](https://github.com/opencv/opencv_zoo/tree/main/models/face_detection_yunet) model
+ * If your version is 4.6 or earlier, download the [2022](https://github.com/opencv/opencv_zoo/blob/088c3571ec70df15100a5e4c26894d95951e92e9/models/face_detection_yunet/face_detection_yunet_2022mar.onnx) version of the model.
+
+The versions are coded in the ofApp::setup() function. Different versions might require editing the code.
+
+### Expected behavior
+
+The camera starts, and a red square will be drawn around the faces found.

examples/computer_vision/yunetFaceDetection/src/main.cpp

@@ -0,0 +1,17 @@
+#include "ofMain.h"
+#include "ofApp.h"
+
+//========================================================================
+int main( ){
+
+	//Use ofGLFWWindowSettings for more options like multi-monitor fullscreen
+	ofGLWindowSettings settings;
+	settings.setSize(640, 480);
+	settings.windowMode = OF_WINDOW; //can also be OF_FULLSCREEN
+
+	auto window = ofCreateWindow(settings);
+
+	ofRunApp(window, std::make_shared<ofApp>());
+	ofRunMainLoop();
+
+}

examples/computer_vision/yunetFaceDetection/src/ofApp.cpp

@@ -0,0 +1,112 @@
+// in ofApp.cpp
+#include "ofApp.h"
+
+//--------------------------------------------------------------
+void ofApp::setup(){
+    ofSetLogLevel(OF_LOG_VERBOSE);
+
+    cout << "OpenCV version: " << cv::getVersionMajor() << "." << cv::getVersionMinor() << "."<< cv::getVersionRevision() << "\n";
+
+    // 1. Setup Live Video
+    camWidth = 640;
+    camHeight = 480;
+    grabber.setup(camWidth, camHeight);
+
+    // 2. Load the YuNet Model via the FaceDetectorYN wrapper
+
+    string modelPath;
+
+    if (cv::getVersionMajor()>=4) {
+        if (cv::getVersionMinor()<=6) {
+            // opencv 4.6.0
+            modelPath = ofToDataPath("face_detection_yunet_2022mar.onnx"); 
+        } else {
+            // opencv 4.9.0 and greater
+            modelPath = ofToDataPath("face_detection_yunet_2023mar.onnx");
+
+        }
+    }
+
+
+
+    //string modelPath = ofToDataPath("yunet.onnx");
+    cv::Size inputSize = cv::Size(320, 320); // The size the network prefers
+    float scoreThreshold = 0.6; // Confidence threshold
+    float nmsThreshold = 0.3;   // Non-Maximum Suppression threshold
+    int topK = 5000;            // Keep top 5000 detections before NMS
+
+    // Call the static 'create' method to initialize the detector
+    detector = cv::FaceDetectorYN::create(
+        modelPath,
+        "", // config file (not needed for this model)
+        inputSize,
+        scoreThreshold,
+        nmsThreshold,
+        topK
+    );
+
+    if (detector.empty()) {
+        ofLogError() << "Failed to load YuNet model. Make sure 'face_detection_yunet_2023mar.onnx' is in bin/data/";
+    } else {
+        ofLogNotice() << "cv::FaceDetectorYN loaded successfully.";
+    }
+}
+
+//--------------------------------------------------------------
+void ofApp::update(){
+    grabber.update();
+
+    if (grabber.isFrameNew()) {
+        // Clear previous detections
+        detectedFaces.clear();
+
+        //
+        // 1. Load pixels from the grabber into our ofxCvColorImage
+        colorImg.setFromPixels(grabber.getPixels());
+
+        // 2. Get the cv::Mat from the wrapper image
+        cv::Mat frame = colorImg.getCvMat();
+
+        // 3. Resize the detector's input.
+        detector->setInputSize(cv::Size(frame.cols, frame.rows));
+
+        // 4. Detect!
+        cv::Mat faces;
+        detector->detect(frame, faces);
+        if (faces.rows > 0) {
+            for (int i = 0; i < faces.rows; i++) {
+                // The format is:
+                // [0-3] = bbox (x, y, w, h)
+                // [4-13] = 10 facial landmarks
+                // [14] = score
+
+                int x = static_cast<int>(faces.at<float>(i, 0));
+                int y = static_cast<int>(faces.at<float>(i, 1));
+                int w = static_cast<int>(faces.at<float>(i, 2));
+                int h = static_cast<int>(faces.at<float>(i, 3));
+
+                detectedFaces.push_back(ofRectangle(x, y, w, h));
+            }
+        }
+    }
+}
+
+//--------------------------------------------------------------
+void ofApp::draw(){
+    ofSetColor(255);
+    grabber.draw(0, 0); // Draw the live video
+
+    // Draw the detected faces
+    ofNoFill();
+    ofSetLineWidth(2);
+    ofSetColor(ofColor::red);
+
+    for (auto& rect : detectedFaces) {
+        ofDrawRectangle(rect);
+    }
+
+    // Draw info
+    ofSetColor(ofColor::white);
+    ofDrawBitmapString("FPS: " + ofToString(ofGetFrameRate(), 0), 20, 30);
+    ofDrawBitmapString("Detections: " + ofToString(detectedFaces.size()), 20, 50);
+}

examples/computer_vision/yunetFaceDetection/src/ofApp.h

@@ -0,0 +1,35 @@
+// in ofApp.h
+#pragma once
+
+#include "ofMain.h"
+#include "ofxOpenCv.h"
+#include "ofVideoGrabber.h"
+
+// include the dnn and objdetect headers
+#include "opencv2/objdetect.hpp"
+#include "opencv2/imgproc.hpp"
+
+class ofApp : public ofBaseApp{
+
+public:
+    void setup();
+    void update();
+    void draw();
+
+    ofVideoGrabber grabber;
+    int camWidth;
+    int camHeight;
+
+
+    //
+    // We use a Ptr (a smart pointer) to hold the detector
+    cv::Ptr<cv::FaceDetectorYN> detector;
+    //
+    // -----------------------------
+    //
+
+    // Vector to store the final, processed face rectangles
+    std::vector<ofRectangle> detectedFaces;
+
+	ofxCvColorImage colorImg;
+};