Next, we will see how to modify the face_detection_opencv_dnn.py script (from Chapter 11, Face Detection, Tracking, and Recognition) in order to do the following:
- Compute the output when several images (with possible different sizes) are fed to the network – the face_detection_opencv_cnn_images.py script
- Compute the output when several images (with possible different sizes) are fed to the network when the crop=True parameter in the cv2.dnn.blobFromImages() function—the face_detection_opencv_cnn_images_crop.py script
The output of the face_detection_opencv_cnn_images.py script is shown in the following screenshot:
The output of the face_detection_opencv_cnn_images_crop.py script is shown in the following screenshot:
In the previous screenshot, you can clearly see the difference when the images have been cropped from the center.