Visit LEDU token page

Image Retrieval by Similarity using Tensorflow and Keras


Intro - Image Retrieval by Similarity using Tensorflow and Keras

with nitinsethi|4 years experience


Watch Full Video | +150 LEDU

Image Retrieval by Similarity using Tensorflow and Keras

  • English
  • Artificial Intelligence
  • Computer VisionComputer Vision
  • (7484)
  • Project length: 2h 44m

Ever thought of creating your own face verification system? This project will teach you how to find similar images using similarity. These are the basics of face verification system. In the process you will also learn tensorflow and keras basics (build a linear regression model using tensorflow).


Image Retrieval by Similarity using Tensorflow and Keras

This tutorial will cover all the details (resources, tools, languages etc) that are necessary for image retrieval. You will be guided through all the steps and concepts, starting from the basic ones like setting up the right tools and frameworks to the more advanced topics related to the development.

What are the requirements?

  • Python basics
  • Tensorflow-Keras basics (will be covered in lesson too)
  • Basic neural network concepts
  • And the most important is: desire to learn

What is the target audience?

  • Learners who want to enhance their knowledge
  • This course will help the students who are doing their final projects

Project Outline

Session 1: Tensorflow basics

  • Introduction
  • Variables,Placeholder and tensorflow session
  • Linear regression using tensorflow
  • Training Linear regression using tensorflow
  • Visualizing our linear regression equation

Session 2: Learning Keras through Classification on Mnist dataset

  • Intro to Mnist and one-hot encoding
  • Keras layers
  • Google Colab, model training and hyperparameters

Session 3: Triplet Loss

  • Triplet Loss Introduction
  • Triplet Loss Coding

Session 4: Image Retrieval by Similarity Theory

  • Representation of an image in computer
  • Learning from little data.
  • Image verification and image recognition
  • Challenge in image recognition
  • Siamese Network (One-shot learning)

Session 5: Image Retrieval by Similarity Coding

  • Image: open,read,resize
  • Coding function to get image encodings.
  • Similarity function
  • Data visualization and image encodings
  • Similarity matrix
  • Data preparation for
  • Creating encodings for our dataase
  • Image retrieval
  • Increasing efficiency of this algorithm.

Thank You


Average rating

5(7484 Reviews)