In this chapter, we first gave a quick overview of different deep-learning-based object detection methods. Then we covered in detail how to use the TensorFlow Object Detection API to make off-the-shelf inference with pre-trained models and how to retrain pre-trained TensorFlow Object Detection models in Python. We also provided detailed tutorials on how to manually build the TensorFlow iOS library, create a new iOS app using the library, and how to use the pre-existing and retrained SSD MobileNet and Faster RCNN models in iOS. Finally, we showed what it takes to use another powerful object detection model, YOLO2, in your iOS app.
In the next chapter, our third computer-vision-related task, we’ll take a closer look at how to train ...
