Move preprocessing out of main

examples/inception_v3/CMakeLists.txt

@@ -6,6 +6,7 @@ project(Inceptionv3
 
 add_library(inception_v3_lib 
     inception_modules.cpp
+    inception_utils.cpp
 )
 
 find_library(CUDANet_LIBRARY NAMES CUDANet HINTS ${CMAKE_CURRENT_SOURCE_DIR}/../../build)

examples/inception_v3/inception_utils.cpp

@@ -0,0 +1,46 @@
+#include <iostream>
+#include <opencv2/opencv.hpp>
+#include <vector>
+
+std::vector<float>
+readAndNormalizeImage(const std::string &imagePath, int resizeSize, int cropSize) {
+    // Read the image using OpenCV
+    cv::Mat image = cv::imread(imagePath, cv::IMREAD_COLOR);
+    // Convert the image from BGR to RGB
+    cv::cvtColor(image, image, cv::COLOR_BGR2RGB);
+
+    // Calculate the scaling factor
+    double scale = std::max(static_cast<double>(resizeSize) / image.cols, static_cast<double>(resizeSize) / image.rows);
+
+    // Resize the image
+    cv::Mat resized;
+    cv::resize(image, resized, cv::Size(), scale, scale, cv::INTER_AREA);
+
+    // Calculate the cropping coordinates
+    int x = (resized.cols - cropSize) / 2;
+    int y = (resized.rows - cropSize) / 2;
+
+    // Perform center cropping
+    cv::Rect roi(x, y, cropSize, cropSize);
+    image = resized(roi);
+
+    // Normalize the image
+    image.convertTo(image, CV_32FC3, 1.0 / 255.0);
+
+    cv::Mat mean(image.size(), CV_32FC3, cv::Scalar(0.485, 0.456, 0.406));
+    cv::Mat std(image.size(), CV_32FC3, cv::Scalar(0.229, 0.224, 0.225));
+    cv::subtract(image, mean, image);
+    cv::divide(image, std, image);
+
+    // Convert the 3D image matrix to a 1D array of floats
+    std::vector<float> imageData;
+    for (int c = 0; c < image.channels(); ++c) {
+        for (int i = 0; i < image.rows; ++i) {
+            for (int j = 0; j < image.cols; ++j) {
+                imageData.push_back(image.at<cv::Vec3f>(i, j)[c]);
+            }
+        }
+    }
+
+    return imageData;
+}

examples/inception_v3/inception_v3.cpp

@@ -3,34 +3,8 @@
 #include <cudanet.cuh>
 #include <iostream>
 #include <opencv2/opencv.hpp>
+#include <vector>
 
-std::vector<float>
-readAndNormalizeImage(const std::string &imagePath, int width, int height) {
-    // Read the image using OpenCV
-    cv::Mat image = cv::imread(imagePath, cv::IMREAD_COLOR);
-
-    // Resize and normalize the image
-    cv::resize(image, image, cv::Size(width, height));
-    image.convertTo(image, CV_32FC3, 1.0 / 255.0);
-
-    // Normalize the image https://pytorch.org/hub/pytorch_vision_alexnet/
-    cv::Mat mean(image.size(), CV_32FC3, cv::Scalar(0.485, 0.456, 0.406));
-    cv::Mat std(image.size(), CV_32FC3, cv::Scalar(0.229, 0.224, 0.225));
-    cv::subtract(image, mean, image);
-    cv::divide(image, std, image);
-
-    // Convert the 3D image matrix to a 1D array of floats
-    std::vector<float> imageData;
-    for (int c = 0; c < image.channels(); ++c) {
-        for (int i = 0; i < image.rows; ++i) {
-            for (int j = 0; j < image.cols; ++j) {
-                imageData.push_back(image.at<cv::Vec3f>(i, j)[c]);
-            }
-        }
-    }
-
-    return imageData;
-}
 
 int main(int argc, const char *const argv[]) {
     if (argc != 3) {
@@ -55,7 +29,7 @@ int main(int argc, const char *const argv[]) {
     inception_v3->loadWeights(modelWeightsPath);
 
     std::vector<float> imageData =
-        readAndNormalizeImage(imagePath, inputSize.first, inputSize.second);
+        readAndNormalizeImage(imagePath, inputSize.first, inputSize.first);
 
     // Print the size of the image data
     const float *output = inception_v3->predict(imageData.data());

examples/inception_v3/inception_v3.hpp

@@ -2,6 +2,10 @@
 #define INCEPTION_V3_HPP
 
 #include <cudanet.cuh>
+#include <vector>
+
+std::vector<float>
+readAndNormalizeImage(const std::string &imagePath, int resizeSize, int cropSize);
 
 class BasicConv2d : public CUDANet::Module {
   public: