Member-only story

Image Similarity Detection in Action with Tensorflow 2.0

Ready to use pipeline for your web application

Erdem Isbilen
TDS Archive
10 min readDec 29, 2019

--

Photo by Sharon McCutcheon on Unsplash

In this post, I will show you how I implemented the ‘Image Similarity Detection’ task in my ‘Fashion Price Comparison’ web application. I will use image similarity to suggest users visually similar products based on what they searched.

The full source code of the implementation is available in my GitHub repository.

Throughout the post, there will be dedicated sections for each one of the following subjects;

  • How to use Tensorflow 2.0 and Tensorflow Hub to generate ‘image feature vectors’ of the product images.
  • How to use Spotify/annoy library and image feature vectors to calculate the image similarity scores.
  • Storing similarity scores and related product identification numbers in a JSON file to enable visual search in our web application.

What Is ‘Image Similarity Detection’ and Why It Is Important?

Image similarity detection is used to quantify the degree of visual and semantic similarity of the images.

Duplicate product detection, image clustering, visual search, and recommendation tasks are performed with this technology in…

--

--

TDS Archive
TDS Archive

Published in TDS Archive

An archive of data science, data analytics, data engineering, machine learning, and artificial intelligence writing from the former Towards Data Science Medium publication.

Erdem Isbilen
Erdem Isbilen

Written by Erdem Isbilen

Machine Learning and Data Science Enthusiasts, Automotive Engineer, Mechanical Engineer, https://www.linkedin.com/in/erdem-isbilen/

Responses (6)