Letâ€™s learn a new phrase today: Supervised Multiclass Labeling, or SML. Itâ€™s a new kind of search engine for images, a new algorithm to improve image labeling, and a quicker method of finding the images you want.
For example, instead of getting photos of a Tiger MPX motherboard, Tamiya Tiger toy tank, Tiger Woods, Purple Tiger flower, and many other things when searching for â€œtiger,â€ you get only photos of a big cat with stripes. SML analyzes the images themselves rather than looking at captions or other text linked to the images.
Engineers at the University of California San Diego (UCSD) are developing the new image search engine which they say can be used to annotate and search commercial and private image collections.
At the core of the SML system is a set of simple yet powerful algorithms developed at UCSD. Once you train the system, you can set it loose to recognize and label images on a database. Once an image is labeled, it can be retrieved via keyword searches.