Runway Palette selected as a successful AI Deep Learning application

Runway Palette [Photo provided = Runway Palette]

AI technology is expanding its realm. Algorithm and Deep Learning facilitated indexing of needed data through extraction from massively collected big data.

Runway Palette is an application that helps finding data through indexing color.

Runway Palette is a project that combined AI algorithms with Google Arts & Culture. It indexes fashion show photos by color. Colors may be chosen and pictures of fashion shows consisting of similar colors is shown.

Google received 140000 fashion show photos from the fashion media The Business of Fashion (BOF) and used AI algorithm called Machine Learning to classify color data.

Machine Learning is an algorithm that parses data learns from it and applies what it has learned to make decisions. The technology aims to realize functions such as human learning capabilities on a computer.

The project is completed with a huge palette of colors and 1000 styles 140000 photos taken at 4000 fashion shows may be covered and searched at once.

It is possible to check what colors were popular by season and era and what design trends were like. In addition colors designers seasons and trends at fashion weeks around the world may be identified.

Runway Palette considered to be one of the most successful AI Deep Learning applications provides similar search results based on analysis. This is realized by finding query image features. More than 1000 designers participated in the experiment for four years using fashion show data from 2015 to 2019.

☞link : https://artsexperiments.withgoogle.com/runwaypalette

By accessing the link above and clicking on a color palette consisted of numerous colored dots the color will direct to related fashion show photo list. The color palette is visualized in 2D and 3D and when the thumbnail of the photo is clicked it displays information such as the name of designers and fashion shows and the period of the fashion show held.

Website: : http://www.wip-news.com/news/articleView.html?idxno=7477