Deep learning for image segmentation using Tensorflow 2
Deep learning for image segmentation using Tensorflow 2
Duration: 4h 43m | .MP4 1280x720, 30 fps(r) | AAC, 44100 Hz, 2ch | 2.34 GB
Genre: eLearning | Language: English
Duration: 4h 43m | .MP4 1280x720, 30 fps(r) | AAC, 44100 Hz, 2ch | 2.34 GB
Genre: eLearning | Language: English
Train and evaluate Mask RCNN instance segmentation model | Train locally and on google ai platform for custom datasets
What you'll learn
You will learn what image segmentation is in computer vision
You will learn what is the difference between instance segmentation and semantic segmentation
You will learn how Mask RCNN deep learning model works
You will learn how to prepare a custom dataset from scratch for your training
You will learn how to annotate your dataset using open source tools
You will learn how to transform your images and annotations into tfrecord format
You will learn how to install Tensorflow 2 with GPU support (Linux and Windows)
You will learn how to install Tensorflow 2 object detection API (Linux and Windows)
You will learn how to run the training of Mask RCNN model on your local machine (Linux and Windows)
You will learn how to create a google cloud account
You will learn how to setup a project on google cloud platform
You will learn how to upload your data to google storage using buckets
You will learn how to run your training of Mask RCNN model on google ai platform
You will learn how to run the evaluation of Mask RCNN model on google ai platform
You will learn how to export your SavedModel (production model) from your training checkpoints
You will learn how to use your production ready Mask RCNN model to perform image segmentation on new images
Requirements
Basic understanding of Python (you should know what functions are and how to use them in Python)
Basic understanding of deep learning (you should know what a neural network is and what training is)
Description
This course is about using deep learning to perform image segmentation with Tensorflow 2. It will show you a step by step guide on how to build powerful deep learning driven image segmentation tasks in computer vision.
The course will show you how to use Mask RCNN deep learning model in order to perform image segmentation. Mask RCNN is one of the widely used neural networks for image segmentation tasks.
The course will help you answer these questions:
1/ What is image segmentation?
2/ What are the different types of segmentation in computer vision?
3/ How do you prepare a custom dataset for training Mask RCNN model?
4/ What tools are used for annotating a dataset for image segmentation?
5/ How do you transform your images and annotations to tfrecords format?
6/ How do you use Tensorflow 2 object detection API for training Mask RCNN model?
7/ How do you use Tensorflow 2 object detection API for evaluating Mask RCNN model?
8/ How to run the training of Mask RCNN model on your local machine?
9/ How to create an account on google cloud platform (GCP)
10/ How to setup a project on google cloud platform (GCP)
11/ How to run the training of Mask RCNN model on google ai platform?
12/ How do you export a SavedModel from your training checkpoints?
13/ How do you use your SavedModel to perform image segmentation on new images?
14/ How do you use Mask RCNN to build a powerful image segmentation model for segmenting different parts of a damaged car (door, hood, lamps, …). Which is by the way the course project!
And a lot more!
My strategy with this course is to enable you to build powerful AI solutions for image segmentation in computer vision.
Who this course is for:
Students
DIY makers
AI Hobbyists
Machine learning enthusiats
Machine learning engineers
Computer vision enthusiasts
Computer vision engineers
Data scientists
More Info