Рет қаралды 6,047
Code generated in the video can be downloaded from here:
github.com/bns...
In this video, we will create a content-based image retrieval system which is basically an image-based search. We achieve this task by storing features from images into a database that we will search to retrieve images. Features can be generated many ways. In this tutorial I will extract custom features using a few digital image filters. I will also show feature extraction using pre-trained VGG16 and ResNet50 networks on Imagenet database.
You will learn about the importance of features along the way.
The features from this query image are compared against features from the indexed database and a match score gets reported. The match is performed using the cosine distance method.
en.wikipedia.o...
The top 3 matching image names are then printed on the screen.