Searching images can be a pain at times when you're looking for a specific kind. Google has rolled out an experimental feature dubbed Image Swirl in Google Labs to ease the trouble during image search. As Google wisely puts it -- 'the underlying technology is a work in progress, no specific time could be determined when it will be available by default in Google Search'.
This total eye candy feature for image search literally mocks the Bing's Visual Search function introduced in mid-September.
Google Image Swirl is an experimental image search interface that shows up a grid of 12 thumbnails stacked with images. This grid represents most relevant group of images. On clicking on any of the 12 stacks, the page shows up a big circular cluster connected to a small cluster circle with remaining 11 thumbnails. The effect shown for the thumbnail grid transforming to Cluster circles is referred as "Swirl".
Sub-groups can be further explored in the cluster. The new experimental Images Swirl is a combined implementation of Picasa Face Recognition and Similar Images. At this moment about 2,00,000 queries can be identified by Image Swirl and more will be added over a period of time.
Apparently the image search result clusters will be dependent on the image metadata. With billions of users uploading images to the web without any description or metadata, getting desired images would still require some effort. As image analysis technology advances the image search engine will rely more on the graphical recognition of image than the metadata attached to the same.
This total eye candy feature for image search literally mocks the Bing's Visual Search function introduced in mid-September.
Google Image Swirl is an experimental image search interface that shows up a grid of 12 thumbnails stacked with images. This grid represents most relevant group of images. On clicking on any of the 12 stacks, the page shows up a big circular cluster connected to a small cluster circle with remaining 11 thumbnails. The effect shown for the thumbnail grid transforming to Cluster circles is referred as "Swirl".
Sub-groups can be further explored in the cluster. The new experimental Images Swirl is a combined implementation of Picasa Face Recognition and Similar Images. At this moment about 2,00,000 queries can be identified by Image Swirl and more will be added over a period of time.
Apparently the image search result clusters will be dependent on the image metadata. With billions of users uploading images to the web without any description or metadata, getting desired images would still require some effort. As image analysis technology advances the image search engine will rely more on the graphical recognition of image than the metadata attached to the same.
0 Responses to Google Swirl makes Image Search Easier