Machine Learning for Visual Similarity Search

Early search engines powered by the likes of Yahoo! and Alta Vista offered simple keyword matching technology that helped users find content on the web. These services would count up the number of times a given search term was present on a web page and then rank search results based on keyword frequency. This approach was later augmented by Google and others where search results relevance were improved by analyzing the relationship between different websites. With advancements in machine learning, new techniques have been developed that allow users to search for content without using keywords at all (Yang et al. 2017).

This paper explores how machine learning has been applied to image processing to enable search systems that can use images instead of search terms and return search results based on visual similarity alone.