Python Image Processing Cookbook

How it works...

If you run the code blocks and plot the input and output the segmented images, you will obtain the following figure:

Mask R-CNN can detect 90 classes, including people, animals, vehicles, signs, food, and so on (refer to the object_detection_classes_coco.txt file to see the supported classes).

The cv2.dnn.blobFromImage() function was used to construct a blob from the input image and then pass it to the deep-neural-net model.

The input image needs to be preprocessed before it can be fed to the deep learning model (this was also done by the same cv2.dnn.blobFromImage() function). Preprocessing steps normally involve mean subtraction, ...