Welcome to this 2 hour long guided project on creating and training an Object Localization model with TensorFlow. In this guided project, we are going to use TensorFlow's Keras API to create a convolutional neural network which will be trained to classify as well as localize emojis in images. Localization, in this context, means the position of the emojis in the images. This means that the network will have one input and two outputs. Think of this task as a simpler version of Object Detection. In Object Detection, we might have multiple objects in the input images, and an object detection model predicts the classes as well as bounding boxes for all of those objects. In Object Localization, we are working with the assumption that there is just one object in any given image, and our CNN model will classify and localize that object. Please note that you will need prior programming experience in Python. You will also need familiarity with TensorFlow. This is a practical, hands on guided project for learners who already have theoretical understanding of Neural Networks, Convolutional Neural Networks, and optimization algorithms like Gradient Descent but want to understand how to use use TensorFlow to solve computer vision tasks like Object Localization.
- Create synthetic data for model training
- Create custom metrics and callbacks in Keras
- Create and train a multi output neural network to perform object localization
Welcome to Object Localization with TensorFlow. This is a project-based course which should take less than 2 hours to finish. Before diving into the project, please take a look at the course objectives and structure:
In this course, our primary learning objective is to create and train a multi output convolutional neural network to perform object localization. We will also learn to create custom callbacks and custom metrics in Keras.
This course is divided into 3 parts:
-
Course Overview: This introductory reading material.
-
Object Localization with TensorFlow: This is the hands on project that we will work on in Rhyme.
-
Graded Quiz: This is the final assignment that you need to pass in order to finish the course successfully.
The hands on project on Object Localization with TensorFlow is divided into following tasks:
Task 1: Introduction Task 2: Download and Visualize Data Task 3: Create Examples Task 4: Plot Bounding Boxes Task 5: Data Generator Task 6: Model Task 7: Custom Metric: IoU Task 8: Compile The Model Task 9: Custom Callback Task 10: Model Training
