SEATTLE, WA -- 06/26/08 --
Eyealike, an innovator of
visual-based search platforms for leading Web sites, revealed today new
Attraction Trait Levers for Eyealike Faces, an image-to-image search
solution for facial recognition and similarity matching. With big
intentions to transform the way people search social networks and online
dating websites, Eyealike Faces is based on patent-pending algorithms that
uniquely understand the geometry of the face, including similarities in
skin tone, eyes, hair color, hair texture and length, as well as hundreds
of other relevant data points. The addition of the new Attraction Trait
Levers allows users to further refine their search results to find more
relevant dating prospects based on personal preferences for face shape,
skin color, and hair color.
"Having a physical connection with someone is one of the most important
components of a relationship, yet a huge percentage of first dates never
turn into second dates due to a lack of mutual attraction," says Mark
Brooks a consultant to the online dating industry and Editor of Online
Personals Watch. "Some dating sites will reverse the process so singles can
start with a pool of people they are attracted to and then drill down
further with specific criteria. This will be a much more fun and efficient
means of finding a true match."
There's No Sex Appeal in Checking Boxes: Physical Attraction Matters
Searching for your soul mate is not easy, but in today's dating world, Match.com, AmericanSingles.com, Yahoo! Personals and other sites
open up the possibilities with a few clicks of a mouse and 24-7 access.
These online dating portals facilitate finding potential matches with
text-based user input via keywords and checked boxes, which are not
naturally designed for image-based search and retrieval because they rely
on rich hyperlinks, anchor tags, metadata, path names, font size, and page
formatting to deliver results. Though the criteria may vary slightly by
site, users can typically specify location, age range, education, income
level, marital history, height, body type, eye and hair color, political
leanings, interests, and so on. After sorting by the desired parameters,
comes the manual part of weeding out individuals based on physical
attraction, or lack thereof.
New Visual Search Approach for Dating and Social Networking Sites
Powered by the Eyealike Visual Search Platform, Eyealike Faces takes a new
sophisticated image-to-image search approach that relies on facial
recognition and similarity matching technology over keywords to connect
consumers to relevant images. The technology has been load tested and will
scale to service the largest online dating and social networking sites in
the world.
"Our conversations with multiple dating sites worldwide tells us that
facial recognition and matching technology can help inject the rules of
attraction that take place at a bar, on the ski slopes, or at a friend's
wedding into online dating," said Greg Heuss, president of Eyealike. "With
a unique visual-based approach to search, we can help daters quickly find
people they are attracted to, whether it's someone who looks like their
favorite movie star or anyone else they find attractive."
Following over four years of proprietary research and development
supplemented by unique technologies and expertise from the University of
Washington, the strength of Eyealike Faces lies in its object/facial exact
and similarity matches by extracting and comparing hundreds of features in
an image using smart, 3D model-driven analysis. All together, these
features provide results far closer to what the human eye sees at
unprecedented speed, scale, and relevancy compared to systems that employ a
single method for shape/model analysis.
The new Attraction Trait Levers take online dating even further, allowing
users to quickly find people that look good to them on paper, and in person
simultaneously. For example, a user could upload a photo of someone they
find attractive, whether it is an "ex," a friend or even a celebrity. If
they found Jennifer Anniston beautiful, but had a preference for short red
hair and a round jaw vs. the Hollywood star's blonde hair and pointed
jaw-line, they could set the levers to search the database for relevant
matches. The dater would then sit back while Eyealike Faces processed
detailed criteria to return a set of customized results so they could
immediately start connecting with people based on both physical chemistry
and compatibility via the sites established search criteria.
Pricing, Availability, and Technology
Available immediately, online dating and social networking websites
interested in seeing a demo of Eyealike Faces and discussing how they can
enhance their website with visual-based search should email
faces.demo@eyealike.com. Pricing is based on an annual licensing fee and
varies depending on CPU and storage size, modules purchased, and frequency
of utilization. There's no need to replace current technology or a Website
front end because the platform easily integrates with existing Website
keyword and attribute-based search systems. Eyealike Faces is compatible
with any Windows or Linux server and is optimized to work on Windows and
Mac computers as well as most mobile and smart phones.
About Eyealike
At the forefront of visual-based search, Eyealike offers the first
enterprise-class search platform for facial similarity/recognition, image
detection, and video copyright surveillance. Based on unique patent-pending
technology, the Eyealike Visual Search Platform offers an entirely new
approach that will dramatically advance the way people search, find, and
access people, images, and other rich media content. Following four years
of extensive research and development supplemented by sophisticated
technologies from the University of Washington, Eyealike offers a modular
approach that gives companies the flexibility to meet today and tomorrow's
visual search needs. Visit www.eyealike.com for more
information.
Eyealike and their respective logos are trademarks, registered trademarks,
or service marks of Eyealike. Other product and company names mentioned are
the property of their respective owners.
Image Available: http://www2.marketwire.com/mw/frame_mw?attachid=787403
Media Contacts:
Michele Mehl
425-205-9444
Email Contact
Deanna Leung
206-915-0512
Email Contact
