November 23, 2017 1 min to read

Content Based Retrieval

Content Based Retrieval

ABOUT

Image search engines encompass methods to make a dataset of images visually searchable using only the contents of the image. Academics call this Content-Based Image Retrieval (CBIR). Throughout the rest of this course I will be using the terms “image search engine” and “CBIR” interchangeably.

At the core, image search engines rely on extracting features from images and then comparing the images for similarity based on the extracted feature vectors and distance metric.

Pipeline

• Defining image descriptor
• Feature extraction and indexing dataset
• Defining similarity metric
• Searching

Tools Used

• OpenCV
• Python