Google has today launched Google Image Swirl, a new image discovery tool that visually clusters similar images.
Google Image Swirl starts with a basic image query, then visually matches similar image based on that query. For example, a search for Melbourne clusters day time city shots, night time city shots, and more.
Users then pick the cluster they want, and the images are represented visually is what Google calls a “swirl.” Think circles with other circles and lines (see image above.) Users can then drill down on a new level of clusters that more closely reflect similarities in images until they eventually find the image they are after.
According to Google, “Image Swirl currently works for more than 200,000 queries and we plan to include more queries in the future. Available queries will auto-complete as you start to type in the search box, similar to Google Suggest.”
I’m not particularly fond of visual representations such as what Google is offering here, and plenty have tried similar interfaces in the past and failed miserably. However in actual use, Google Image Swirl is rather nifty. It does make finding the right image easier, be it when you’re after a specific shot or angle vs a simple generic image that you’d find in half the time in Google Image Search.
You can give Google Image Swirl a go at image-swirl.googlelabs.com