Deep Learning: Image Recognition
2h 15mIntermediate2024-08-20
Authors
Isil Berkun
Data Scientist at Intel Corp.
Course details
Deep learning and image recognition is everywhere, from unlocking phones to tagging friends in photos. Learning how it works is critical for anyone working in tech today, especially if you want to stay ahead of the curve, sharpen your skills, and prepare to innovate. Join instructor Isil Berkun as she shows you how to make computers recognize images, how to prepare pictures for AI, and how to build systems that can tell who's who. Along the way, learn about what to avoid and what to do when common problems arise. By the end of this course, you’ll be prepared to build image recognition models and start exploring how to make AI more creative with images.
Learning objectives
Apply the fundamentals of image processing for model training and achieving better performance in image recognition tasks.
Understand the mechanics of CNNs and explore advanced CNN architectures like ResNet and Inception to apply to projects.
Implement a simple image detection model using a pretrained CNN and apply it to different images to understand image classification.
Implement techniques for cleaning, transforming, and feeding data into models for optimal deep learning performance.
Build on the basics of image detection to create systems that can recognize and distinguish individual images.
Select and apply the right metrics to assess model performance and refine models.
Understand the challenges and limitations of current image recognition technologies and ethical implications.
Learning objectives
Apply the fundamentals of image processing for model training and achieving better performance in image recognition tasks.
Understand the mechanics of CNNs and explore advanced CNN architectures like ResNet and Inception to apply to projects.
Implement a simple image detection model using a pretrained CNN and apply it to different images to understand image classification.
Implement techniques for cleaning, transforming, and feeding data into models for optimal deep learning performance.
Build on the basics of image detection to create systems that can recognize and distinguish individual images.
Select and apply the right metrics to assess model performance and refine models.
Understand the challenges and limitations of current image recognition technologies and ethical implications.
Skills covered
Real-TimeNeural Networks and Deep LearningPythonVisualization and Real-TimeAECProduct and ManufacturingArtificial Intelligence (AI)Open SourceDeep Dive (X:Y)
Concepts
0. Introduction
- 01 - Learning image recognition
1. Diving into Codespaces
- 02 - Codespaces - Your new best friend
- 03 - DL Image Recognition libraries with Codespaces
2. Understanding Deep Learning for Images
- 04 - Basics of image processing
- 05 - Convolutional neural networks (CNNs)
- 06 - Advanced CNN architectures
- 07 - Challenge - Simple image classification
- 08 - Solution - Effective image classification techniques
3. Image Recognition Fundmantals
- 09 - Image recognition fundamentals
- 10 - Preprocessing and feeding data into your network
- 11 - Developing image recognition systems
- 12 - Success metrics
- 13 - Challenges in image recognition
- 14 - Challenge - Dealing with noise in images
- 15 - Solution - Dealing with noise in images
- 16 - Generative AI and image recognition
Conclusion
- 17 - Continue your deep learning journey